Subscribe now

Hi friends,

A few weeks ago, I wrote about how AI application companies are increasingly going full-stack, integrating down into the model layer or up into the service layer. Since then, there’s been a lot of discussion around the pros and cons of going into the model layer and when the right time is, which will be the focus of today’s piece.

To be clear, this discussion is in the context of application-layer companies, not frontier labs. Very few of these companies are pre-training from scratch. Instead, most are post-training and RL on strong open-weights bases.

In this piece, I’ll cover:

• The training spectrum

• The case for doing it

• The case against doing it

• When it makes sense, and the new infrastructure that’s lowering the bar

I. The training spectrum

“Training your own model” gets used to describe wildly different commitments.

At the extreme end, there’s prompt engineering, RAG and harness engineering, which isn’t training in any form. A step up, fine-tuning a small model. Further up, supervised fine-tuning or RL on a strong open-weights base as the primary model in the system. Further still, continued pre-training on top of an open-weights model. At the far end, pre-training from scratch.

For application companies, basically nobody is at the far end but as you go right from there, you see examples at the different points. Cursor’s Composer 2 builds on Kimi K2.5. Intercom’s Fin Apex 1.0 sits on an undisclosed open-weights base. Cognition’s SWE-1.5 is, in their words, “end-to-end RL on real task environments using our custom Cascade agent harness on top of a leading open-source base model.”

As Intercom’s CEO put it, pre-training has become “kind of a commodity” and the action is in post-training. And so most of this piece will be in the context of post-training rather than pre-training.

II. The case for training

In my view there are three real reasons to invest in post-training.

1. Unit economics and latency

Once you’re at scale, API calls add up. A smaller specialized model that runs cheaper and returns in 200ms can beat a frontier call that takes 2 seconds and costs 10x as much.

Intercom’s Fin Apex 1.0 reportedly runs at roughly one-fifth the cost of frontier models, responds 0.6 seconds faster than the next-fastest competitor, and resolves customer issues at a higher rate. At the scale of a company like Intercome doing ~2M conversations per week, that gap in latency and unit economics is meaningful.

There’s a related destiny argument: if your unit economics depend on one frontier API, you’re exposed to pricing changes, rate limits, and the provider showing up in your category. No one has felt this more than Cursor, which despite tremendous success and usage, reportedly has -21% gross margins, and has found it difficult to compete with Claude Code when Anthropic’s limits supports thousands of dollars of equivalent compute in their $200 max plan.

2. Differentiation through proprietary data

If everyone is calling the same frontier API, where’s your edge? Increasingly, it has to come from the traces you’ve accumulated.

Cursor sees which completions get accepted and rejected. Intercom has billions of customer-service interactions. OpenEvidence has the queries and citations of 40% of US physicians and trained a domain-specialized model which captured data from peer‑reviewed medical literature via their partnerships. That’s proprietary training data and the best way to put it to work is through training or post-training.

In some ways, the real vaue of the application layer is that they get the best telemetry on the real use cases of their customer base, far better than any off-the-shelf evals can measure.

And by leveraging those to post-train models allows them to actually improve performance for their users while also getting the latency and unit economics benefits outlined above.

For example, Cursor’s Composer 2 was in part optimized on their internal eval Cursor-bench which they created based on their real traces. They felt this benchmark was more representative of the real work developers were doing within their platform than the publicly available benchmarks.

3. Specialized models for the parts frontier labs don’t prioritize

Most application companies aren’t just training one big custom model to replace the frontier. They’re running systems of small specialized models, each fine-tuned for one part of the pipeline that the frontier labs don’t optimize for.

Decagon talks about this: smaller fine-tuned models for query rewriting, routing, and intent classification, with a frontier model only where it’s actually needed. Sierra trained custom search models (Linnaeus and Darwin) inside a constellation that still uses OpenAI, Anthropic as core models. Cognition has SWE-grep for context, SWE-check for bug detection, SWE-1.5 for the main agent.

Most of the value is in the boring parts of the pipeline (voice activity detection, query reformulation, retrieval ranking, tool selection). None of these need a frontier-grade reasoner. All of them benefit from being faster, cheaper, and tuned to your specific data.

This is a great way to step into fine-tuning and post-training for many companies: train where the frontier underserves you and keep using the frontier where it doesn’t.

III. The case against

The biggest reason to be careful: your post-trained model may not survive the next base-model release from the lab. The labs are now releasing new models faster than ever, because the labs themselves are using their own models to build the next ones.

Anthropic’s Dario Amodei has said that 70-90% of the code for new Claude models is now written by Claude itself. OpenAI was even more direct in their GPT-5.3-Codex announcement (now 0.2 generations behind :)):

“GPT-5.3-Codex is our first model that was instrumental in creating itself. The Codex team used early versions to debug its own training, manage its own deployment, and diagnose test results.”

What this means: model releases that used to take months are now arriving weeks apart. OpenAI shipped GPT-5, 5.2, 5.3, 5.4, and 5.5 within months.

For an app company, that’s the biggest risk. A lot of fine-tuning wins from 2022-2024 dissolved when GPT-4 and Claude 3.5 came out, and the cycle is faster now.

That’s why in general, it’s much safer to post-train or fine-tune specialized models that work in a system of models rather than the core reasoning model for frontier tasks since the cost and latency benefits of those are more likely to survive even as base models improve.

There are other costs to consider as well namely that post-training talent is scarce and expensive. The opportunity cost of the talent and capital used to post-train models could be better deployed or leveraged in other aspects of the product and company.

IV. When to do it

One useful proxy is to train when you have enough proprietary traces to make a small specialized model meaningfully better than the frontier on a specific part of your pipeline.

I should note that the bar to start is also lower than it was a year ago, because a new infrastructure layer has shown up to support post-training in different forms.

• Tinker from Thinking Machine Labs is a managed post-training API. They handle distributed training and LoRA infrastructure while you bring data, algorithms, and environments. As Andrej Karpathy’s noted: it lets users keep about 90% of the algorithmic control while removing about 90% of the infrastructure pain.

• Prime Intellect’s Lab is similar, hosted RL training plus an open Environments Hub with hundreds of community-built RL environments.

• Applied Compute, founded by ex-OpenAI researchers, is a more white-glove version, post-training on enterprise proprietary data using RL environments and is one of many companies offering some version of RL-as-a-service.

• Vendors such as Mercor, Surge AI, Fleet and others sell custom expert-authored RL environments

• Lastly, the Chinese labs and their open-source base models that are competitive with the frontier closed source models serve as a starting-off point for many.

I do think the improved infrastructure has meant that even small teams of 10-20 can now post-train if they want to. But one mantra I like that still holds true for most application layer companies is “no GPUs before PMF.” If you don’t have the product or the traces yet, training a model to do that product is premature.

Post-training should be part of the conversation once a company is either rapidly scaling and has collected enough traces or at least has PMF and feels underserved by the frontier in some specialized aspects of their product which they feel the base model can fix.

Closing Thoughts

The companies integrating down into the model layer (Cursor, Intercom, Sierra, Decagon, Cognition, OpenEvidence) aren’t doing it because they like training models. They’re doing it because at their scale, with their traces, the economics and differentiation arguments finally pencil out. And almost all of them are doing it as post-training, not pretraining from scratch.

For most app companies earlier in their lifecycle, the honest answer in 2026 is: not yet, but start setting up to. Build the data collection now (traces, evals). Start with one small specialized model in a boring part of your pipeline rather than trying to replace the frontier on the main reasoning call.

The durable training investment is the data and environments you accumulate, which let you keep producing better models as the bases under you keep improving. And remember, those base models are improving faster than ever.

If you’re working on an AI application or thinking about this tradeoff, feel free to reach out at tanay at wing.vc.

Subscribe now

Hi friends,

I’ve been thinking about a pattern that we’re increasingly starting to see across AI application companies. Over time, I think most of them become “full-stack”.

At a high level, you can think about an AI product that achieves outcomes as having three layers:

1. At the bottom, the model

2. In the middle, the application or agent which includes the data/context, etc

3. At the top, the human or service layer needed to review/prompt/do the last mile to actually get to an outcome

This is obviously a simplified way to describe the world. Models sit on top of chips, etc and there’s a lot of infrastructure, data systems, orchestration, evals in building agents. But I still think the simplification is useful.

Traditional application layer companies would sit just in the middle layer. But these companies are increasingly beginning to (or starting off) vertically integrate in one of two directions. Some move down into the model layer. Others start or move up into the human or service layer. Both end up looking “full-stack1,” just in very different ways.

1. Full stack down

The first version of full-stack is when the application company integrates down into the model layer.

This is the pattern we are seeing in coding, customer service and a few other categories where the company has enough usage, enough proprietary traces, and enough economic incentive to start pulling intelligence in-house.

Cursor is a very recent example. Last week it introduced Composer 2, positioning it as a frontier-level coding model with meaningfully improved benchmark performance and lower pricing than many alternatives. In its technical report, Cursor says Composer 2 uses Kimi K2.5 as a base model, then extends it with continued pretraining plus reinforcement learning on long-horizon coding tasks.

Intercom made a similar move this week with Fin Apex. Intercom CEO Eoghan McCabe described it as “the age of vertical models” and Intercom says Apex now powers essentially all of its English-language chat and email customer conversations.

Cursor, Intercom, Cognition, Harvey, Sierra, and are increasingly trying to shape, tune, route, and in some cases train the intelligence itself around a specific domain.

Why does this happen?

The biggest reason is that for AI companies, this is the most important flywheel they have to drive increased performance. They have the traces of the agents. They see the prompt, the outputs, the edits/acceptances/rejections. Over time, that becomes extremely valuable proprietary training data. Better product creates more usage. More usage creates more traces. More traces make the model better. Then the better model improves the product again.

Other reasons include:

• Cost and Speed: Once you have enough scale, the COGS can add up, and smaller, fine-tuned models can end up giving you enough performance for your use case at much lower cost and at much faster speed.

• Differentiation: If everyone in a category is calling the same handful of models, it becomes harder to build real product distance from the intelligence layer alone.

2. Full stack up

The second version of full-stack goes the other way. Instead of integrating into the model, the company integrates upward into the human or service layer. It sells true outcomes instead of software.

This is the more classical full-stack startup idea that Chris Dixon wrote about in 2015. You do not just sell software into a workflow. You own the end-to-end process and provide an outcome or service to the end customer.

Historically, that often meant ugly services economics and challenges with scaling. But AI changes the shape of that equation and opens up the possibility to take this route in many end markets that weren’t attractive.

This idea of “sell outcomes, not software”, “services-as-software” and AI-native services has been growing in popularity over the last few years, but as models increasingly get better at agentic tasks over longer horizons, we’re now starting to see the first wave of these companies truly break out.

There are a number of examples of these AI-native Services business across various services categories such as legal, insurance, accounting, customer support, recruiting, and IT modernization with the extreme version of these the “roll-up and layer on AI approaches”. On the de-novo AI-native services businesses side, we see companies such as:

• Crosby AI combines software, AI, and attorneys to own more of the delivered result and outcome in what co-founder and CEO Ryan calls a “Neofirm”.

• WithCoverage and Harper building AI-native insurance brokerages

• Mechanical Orchard is an AI-native software modernization services firm

AI is great but it isn’t 100% there, and a company that owns the end result can both plug the gaps in all the missing places is both incentivized to contuosly improve and allows the customer to not have to think about the last mile.

This path is especially interesting because it may eventually absorb even more of the stack. What starts as application plus service can, over time likely post-train specialized models too. As AI does more of the work, some of these companies may end up owning two layers at first, then increasingly all three.

3. Closing thoughts

I think AI application companies will not tend to stay just AI application companies. Over time, they will vertically integrate. Many will use their traces to own the intelligence as well to increase differentiation, performance and reduce costs, particularly given that the model companies themselves are encroaching on some of their applications. The others will either start out delivering full outcomes or over time as they can do an increasing amount of the work, decide they can capture more value by selling the full service rather than an agent that does 90% of it.

For more related to this topic, here are some good pieces:

• Ryan Daniels (Crosby AI) on the Rise of the Neofirm

• Cursor’s Composer 2 Technical Report

• Eoghan McCabe (Intercom) on the Age of Verticalized Models

• Chris Dixon’s original piece on Full-stack Startups

Subscribe now

Hi friends,

We’re currently in the SaaSpocalypse. People believe software is dead and margins will compress to zero. Some are even saying that companies like Visa get bypassed and DoorDash gets aggregated away in the age of AI. Everything that looks like software becomes a commodity and no moats remain.

Before we declare the end of defensibility of all businesses, I think it’s worth grounding ourselves in the actual sources of defensibility that exist. My favourite book around defensibility and moats is Hamilton Helmer’s 7 Powers which outlines the common ways companies build defensibility.

The question is: In an AI world, which sources of power weaken, and which survive? Let’s walk through all seven, particularly in the context of software and technology companies.

1. Scale Economies

What it is

Scale Economies exist when larger volume reduces unit cost in a way smaller competitors simply cannot match, creating a cost advantage.

In traditional software and digital businesses, scale allowed for spreading investments in engineering, infrastructure, and sales across a large customer base, creating structurally lower cost per unit.

For example:

• Amazon Web Services spreads massive capital investments in data centers and infrastructure across millions of customers

• Netflix amortizes its 10B+ content budgets across over 200+ million subscribers, allowing it to have more volume of content and more niche content that competitors.

• Large SaaS vendors like Salesforce historically spread R&D and support costs across thousands of enterprise customers, reinforcing margin and reinvestment advantages.

What AI does

AI compresses labor-based scale advantages in software / digital work. A 20-person team equipped with agents can now build features, handle support, write documentation, and run experiments at a velocity that previously required much larger organizations.

However, at the infrastructure and model layer, scale continues to be important and a source of power.

Net result

Application-layer scale advantages that were based on spreading R&D and similar costs that across large user bases weakens.

Infrastructure-layer scale advantages still remain. In fact, Scale economies provide a source of power for model layer companies like OpenAI and Anthropic.

2. Network Economies

What it is

Network Economies arise when a product becomes more valuable as more participants join. There are two primary types:

• Same-sided network effects, where users benefit directly from other users being present as is the case on social networks and messaging apps.

• Cross-sided networks, where two distinct user groups create mutual value as is common in marketplaces like Uber and Doordash.

For example:

• WhatsApp became indispensable because every additional user increased the utility of the network for everyone else.

• DoorDash strengthened as more restaurants joined, which attracted more consumers, which in turn attracted more restaurants, reinforcing marketplace liquidity.

In each case, value scaled with growth of the network.

What AI does

Agents introduce frictionless multi-homing and can make it easier to simulate aggregating one side of a marketplace even if they didn’t exist. For example, a new food ordering service could use voice agents to call a restaurant not on their platform to order from them. Or a consumer-facing agent can arbitrage across marketplaces to find the best price.

That can weaken shallow exclusivity.

But AI cannot fabricate real-time liquidity, courier density, reputation history, or a canonical identity graph. Marketplace density and trust are structural, not labor-based.

Net result

Some surface-level network effects can weaken but deep liquidity networks with trust, reputation, and coordination density can still remain durable.

3. Counter-Positioning

What it is

Counter-Positioning occurs when a new entrant adopts a business model that incumbents cannot replicate without damaging their existing business. This often happens during technological shifts where new entrants “turn” the incumbents existing assets and business into a weakness.

For instance:

• Netflix’s transition from DVDs to streaming undercut traditional rental economics and forced incumbents into painful strategic trade-offs.

• SaaS companies offering subscription pricing challenged on-prem vendors reliant on upfront license revenue.

What AI does

AI creates new waves of counter-positioning. Startups can offer forms of usage outcome-based pricing instead of per-seat pricing. They can replace workflows with agents rather than augment users.

Incumbents may struggle to adopt these models because doing so cannibalizes existing revenue streams or disrupts organizational incentives.

Net result

Counter-Positioning is a powerful form of power for startups to get going in this current market. However, incumbents are more savvy than ever before and recognize that cannibalization remains needed.

In addition, startups may not be able to counterposition against all players vying to capture the new market such as AI labs, etc who are not tied to prior business models like incumbents.

In aggregate, counter-positioning remains a good source of power early on to get going.

4. Switching Costs

What it is

Switching Costs arise when customers face meaningful pain in leaving a product. In enterprise software, this historically included:

• Complex data migrations

• Rebuilding custom integrations

• Retraining teams on new workflows and systems/UI

It was arguably the primary source of power many software companies seeked to create.

For example:

• Salesforce embeds itself deeply into sales processes, capturing years and years of data, making migration costly and risky.

• Workday runs payroll and HR compliance systems that are tightly integrated into regulatory reporting.

What AI does

AI directly attacks labor-based switching costs. Agents can map schemas, rewrite integrations, generate training materials, and even run systems in parallel to reduce migration risk. What once required months of consultants may compress to weeks of automated orchestration.

In addition, if agents do the workflows, human workflows may not be as relevant and human AIs may evolve to simpler natural language or change regardless, meaning that workflow as a switching cost may also drastically reduce.

Net result

Switching costs for software businesses weaken meaningfully. Companies can more easily adopt new software, get their data in, customize UIs bespoke to their workflows quickly and relatively cheaply.

Some levels of risk remain as a source of friction that retains some level of switching costs, but most companies will no longer feel held hostage by vendors.

If you don’t yet receive Tanay’s newsletter in your email inbox, please join the 10,000+ subscribers who do:

Subscribe now

5. Branding

What it is

Brand reduces evaluation cost. It acts as a shortcut for trust, quality, and reliability.

In enterprise markets, brand often signaled safety. No one got fired for choosing IBM or Microsoft. In consumer markets, brand shaped preference and habit in a noisy world full of unclear claims.

Examples include:

• Microsoft, whose enterprise brand reduces procurement friction across software categories.

• Consumer brands like Nike, where emotional attachment drives durable preference.

What AI does

If agents continuously benchmark products on performance and cost and can do deep research based on individual needs of users/businesses, brand as a heuristic shortcut weakens. Evaluation can become systematic rather than reputational. This may be particularly true for consumer brands where performance and functionality matter more and brand was a shortcut for that than say brands perceived as luxury.

At the same time, AI also introduces new forms of risk: model unpredictability, hallucinations, security concerns. In high-liability contexts, institutional trust becomes more valuable.

This leads to the notion of brand splitting:

• The strength of marketing-driven brand compresses in areas where price/performance tradeoffs matters and agents can thoroughly evaluate that

• Institutional reliability brand persists and may even strengthen in a world where stakes remain high and risks associated with AI could even increase.

Net result

Some forms of brand weaken significantly since deep and thorough evaluations become possible, even for purchases where they didn’t make sense prior.

Other forms of brands persist or even strengthen as trust and accountability become even more important.

6. Cornered Resource

What it is

Cornered Resource exists when a company controls an asset that competitors cannot access.

This may include exclusive data, regulatory licenses, distribution control, or proprietary intellectual property such as patents in biotech/pharma. In software, the most common form of cornered resource has been proprietary data often collected and aggregated via customers directly over long periods of time.

Examples include:

• S&P Global / Moody’s / Fitch have cornered resources in the form of credit ratings, indices, and reference data that are written directly into regulations, covenants, and mandates

• CoStar in commercial real estate has a hard‑won proprietary dataset of listings, comps, historical deals, building attributes, and contacts that brokers and landlords rely on

• Helmer uses the example of Pixar which had a once-in-a-generation cluster of creative talent as a cornered resource

What AI does

AI increases the value of proprietary data because models improve with exclusive signal. At the same time, especially for datasets that were public in some form, AI means that they are no longer “cornered” since they can now be acquired much more cheaply than in the past by leveraging LLMs to scrape / structure those datasets.

So if exclusivity is real and structural, AI strengthens this power. But for some companies that thought they had proprietary data (e.g., data businesses where data was available online) which were not truly proprietary, it shines a light on them not having as much defensibility as they thought.

Net result

True forms of Cornered Resource becomes even more important in an AI-native world. Proprietary data (or rails) and what is enabled with that increases.

7. Process Power

What it is

Process Power arises from deeply embedded organizational routines that compound over time giving companies a durable advantage over in areas such as speed, product quality, cost advantages, etc.

For example:

• Toyota’s Production System is the prototypical examples that allowed allowed them to have manufacturing lines that improved continuously and produced cars at extremely low defect rates.

  

• Netflix’s data-driven commissioning process allows it to greenlight content with confidence.

• Meta’s large scale product experimentation and growth process that allows for continuous improvement of their products for the metrics they care about via large scale experimentation and A/B testing that was iterated on over a decade.

These routines create compounding advantage and provide a source of differentiation over others.

What AI does

AI commoditizes some forms of process advantages that happen in the digital world (product development, etc) because it allows companies to iterate faster than before and allow agents to iterate on their behalf (which arguably have best practices and experience in them via training).

However, data that is proprietary to a company does not become available to competitors, so any source of process power that depends on using that data remains.

At a baseline, the core processes that are digital oriented in any business likely improve because of agents, making it more difficult for competitors to have a delta over them. But TSMC’s process power for example still remains.

Net result

For the companies that had true process power based on 1p data / institutional knowledge / compounding intelligence, their process power will remain but potentially weaken as the competitors may improve. At the same time, by integrating AI into their feedback loops, they could drive even further improvements in their process, and so can stay ahead.

Generic process advantages will weaken.

Closing Thoughts

The powers and what happens to them are summarized above. If you’re building an AI-native company, the reality is that default software moat over the last few decades, switching costs, is likely not sustainable. This may actually help you replace incumbents in that it may be easier to migrate customers over, etc, but it hurts your ability to remain sticky.

In this world, defensibility shifts toward cornered resources (typically proprietary data), real network effects (when your line of business supports them), and compounding process power (being extremely AI-native is a good one to start with). Over time you can also built a trust-based brand which also will remain a source of differentiation.

If you’re interested in discussing further or have thoughts or feedback, I’d love to hear from you!

Subscribe now

Hi friends,

Over the last year, we’ve clearly entered a new era where models improve less from static human data and more from real experience. By “experience” I mean these models taking actions in environments, observing outcomes, getting feedback, and then getting better. When you look at what the frontier labs are doing, it increasingly seems like scaling reinforcement learning is how that experience gets converted into improvement.

In this post, I’ll discuss this new paradigm, the emerging stack around it and the open questions at play.

1. The Era of Experience

We were in the “human data” era where models learned from tokens from both the internet and expert data specialists. We trained huge models on massive datasets, and a lot of capabilities emerged from learning over large amounts of language and code.

Now, in the “experience” era, models learn from attempting tasks and receiving feedback in environments that mimic real tasks that we as humans may face. A model is taking a sequence of actions across tools and time.

Silver and Sutton of DeepMind, in their “Era of Experience” paper describe it as: the next generation of agents will learn from streams of interaction, take grounded actions in real action spaces, receive grounded rewards from consequences, and increasingly plan in the currency of experience.

Perhaps the clearest evidence of this is the shift in how labs allocate compute and spend on data. On the compute side, Dan Roberts, an OpenAI researcher, talks about compute going towards RL scaling and talks about a potential world where compute on RL scaling continues to grow to even beyond pre-training.

We added some reinforcement learning compute for o1 … o3 maybe had a little bit more RL compute. At some point in the future maybe we’ll have a lot of RL compute and then at some far point in the future maybe we’ll have just like be totally dominated and crushed by RL compute - Dan Roberts

Similarly, on the data side, the labs have put over dozens of vendors in business in the last year who sell environments to the labs that mimic real world tasks of some form that can be used to improve model performance on these tasks through RL. More on that later in the piece.

Additionally, Brendan Foody, CEO of Mercor notes that RL is getting so good that it can saturate any eval.

RL is becoming so effective that models will be able to saturate any evaluation. This means that the primary barrier to applying agents to the entire economy is building evals for everything. However, AI labs are facing a dire shortage of relevant evaluations. Academic evaluations that labs goal on don’t reflect what consumers and enterprises demand in the economy.

Continued progress through this then hinges on continuing to have new sets of tasks packaged in the right environments that can be used for the models to “play” in and learn from experience in. An environment is the sandbox where the agent can act (100s of thousands of times) and receive feedback (in the form of rewards). In practice, it’s an instrumented version of a real workflow: a UI/application, an API surface or a codebase.

2. The RL “stack”

As RL continues to scale, first inside frontier labs and over time across enterprises and ISVs, a startup ecosystem has formed in a few concrete areas. At a high level, there are three categories of companies: Environment builders (who supply the “data”), RL infrastructure (infrastructure needed to run different RL jobs) and RL-as-a-service (serving those that want more reliable agents but lack the expertise). Let’s go deeper into each!

RL environment builders

These companies supply the form of data that matters for RL, particularly to the labs: environments where agents can take actions, observe outcomes, and get scored.

In practice, an “environment company” usually sells environments packaged as: a workflow surface the agent can act in (UI, API, codebase, toolchain), a task library so you can generate many attempts, an eval that scores the attempts, and logging so you keep full trajectories.

Most of these companies primarily sell to labs today, and tend to have an area of focus although over time will likely expand areas and also potentially sell to enterprises. Common focuses and examples include:

• Coding: Preference Model, Proximal, Mechanize

• Computer Use: Fleet, Deeptune, Habitat, Matrices

• Other forms of specialized knowledge work: Phinity (Chip design), Isidor (finance)

Many of these companies are scaling revenue rapidly, though it’ll be interesting to see which ones can continue to scale and remain flexible and adaptable over time and needs from labs evolve as well as actually provide environments that mimic real-world scenarios at the right level of difficult to drive model performance.

The “incumbent” data vendors to labs such as Turing, Mercor and Surge are also supplying environments, and indeed arguably have the highest share of the market, in addition to traditional forms of human expert data.

RL infrastructure

While the labs may have their own infrastructure built to manage RL, there are a number of companies that have built various forms of infrastructure to make aspects of RL simpler.

Teams that want to run their own RL loops need tooling for rollouts, data collection, evaluation, policy updates, gating, and iteration. They also need the underlying compute and serving stack to keep throughput high.

Startups building in this area include companies building tooling to simplify RL post-train models such as Miles from RadixArk and Tinker from Thinking Machines to all-in-one compute platforms like Prime Intellect or tooling to make building environments themselves easier such as Hud.

RL-as-a-service

There are many enterprises (and software vendors building agents) that have repeatable workflows for which models and agents don’t work out of the box but could benefit use RL.

To serve this group of companies, a number of startups have emerged that are essentially providing RL expertise and outcomes as a service. The pitch is usually: tell us your workflow / success metric, give us your data (agent traces, workflow data) and we will help do the work needed to post-train the model to give you better outcomes.

Most of these startups have built some level of infrastructure and tooling in-house which they pair with FDEs with RL expertise.

Early customers of these companies tend to be AI-native companies that have a core agent in production that they would like to improve. Over time, the same approach can extend to traditional enterprises for all workflows.

Examples of these companies include Applied Compute, Osmosis, and CGFT.

Applications

So far, most RL scaling has happened inside frontier labs. But given the improving infrastructure, I expect that number of companies to grow, and many AI applications that are sitting on real-world trajectories and outcome feedback in a valuable workflow will likely leverage that.

Coding is a great example of an application category with lots of these trajectories given the adoption. Cursor was able to use the feedback from their tab model to train an improved one using online RL, as they describe in this piece. This is how many application companies will actually see the benefit from the data they collect.

3. Closing thoughts

While we have a new vector of scaling that should continue to result in progress over the short to medium term, there are a number of open questions that remain. For one, how well will the models generalize to pick up new “skills” they are explicitly RL’ed on. Ex-OpenAI research lead Jerry Dworek had this to say:

"How do those models generalize? How do those models perform outside of what they've been trained for? ... How do models do tasks that you don't reinforcement learning on? And probably not that great. And this is... basically the remaining questions in the world of AI because what we train for... we are getting really really good at."

The other is around whether RL scaling enough, or do we need more beyond that, with continual learning being the most obvious gap today. Ilya notes that humans are continual learners and whether we need models to do the same:

"Where on the curve of continual learning is it going to be? ...You produce like a superintelligent 15-year-old that’s very eager to go... they don’t know very much at all... and you say 'go and be a programmer, go and be a doctor, go and learn.' So you could imagine that the deployment itself will involve some kind of a learning trial and error period. It’s a process as opposed to you drop the finished thing."

Regardless of whether RL takes us to AGI however, it’s clearly driving model improvement, and a fruitful area for companies to both build in in the infrastructure area and application companies to leverage over time. If you’re building anywhere in this stack, I’d love to hear from you.

Subscribe now

Hi friends,

ChatGPT is finally starting to look like a real consumer internet product, in the most obvious way: ads. If you’re going to offer a free plan at scale, you can’t subsidize inference forever absent another cash cow business. The question was always when OpenAI would do it, and what the first ad products would look like.

I want to walk through (1) what other AI surfaces have done with ads (2) what we know so far about OpenAI’s approach (3) what the initial opportunity might look like and (4) where this goes once “the user” becomes “the agent.”

Let’s get started.

I. Ads in AI today

OpenAI is by far the most prominent and at-scale surface to start AI surface starts out in ads, but it’s not the first one. A bunch of AI products already monetize various forms of AI surfaces via ads, including:

• Perplexity which has been explicit that advertisers shouldn’t influence answers, and the ad unit they shipped is basically “sponsored follow-up questions” plus paid media alongside an answer.

  

• OpenEvidence in medicine where intent is unusually high and budgets are concentrated (pharma, devices). They’re free to doctors and ad-supported, and they’ve been public about ads being a meaningful revenue driver, reportedly close to / in the 9 figures of revenue.

• Amp Free from Amp Code which offers a free agentic coding experience that is ad supported.

• Talkie (owned by Chinese AI lab Minimax) is a Character.AI style app where people can chat with different characters and generates revenue primarily through advertising (on the free tier) in the form of banners and interstitials and in-app purchases. Talkie notes that their eCPMs range from $2-$10.

The way to think about these early AI ad systems is simple: they’re still advertising to humans. The ad is shown to a person, the person clicks (or doesn’t), and measurement is ultimately downstream conversion and lift. The “AI” part mostly changes targeting (more intent signals because of clear queries/intent) and the canvas (the answer box can absorb a lot of context).

II. OpenAI’s Ads Plans

From what’s been reported so far, OpenAI is taking two approaches to monetize their free tier and capture more of the value they create:

1) Classic intent-based ad formats

The most obvious format OpenAI is starting with is intent-based ads within select conversation as part of the response but clearly separated from the LLM response). OpenAI has also been careful to communicate this fact that ads shouldn’t change the actual answers of the LLM and that they’ll be clearly labeled.

These ads will look more similar to Google’s ads in that targeting will be based on the intent of the queries themselves, but interestingly seem to have a component where you can click through to chat with the Business AI (the action upon click is more similar to “click to chat” ads on Meta). It remains to be seen if users and advertisers will prefer that to clicking through to the website.

The early chatter suggests high CPMs in the $60 range which is shown below (and pricing based on impressions rather than clicks or other forms of downstream performance), limited inventory, and non-self-serve buying to start, which is exactly how you’d expect this to launch. If you’re an advertiser, you’re paying for novelty, perceived intent, and scarcity, and in being early to shape the advertising product to what the ideal is over time (various ad formats, measurement, targeting, brand safety, etc). My guess is these premium CPM driven ads don’t scale and ChatGPT adopts a more classic performance-based model driving real results (leads, conversions, clicks) and charging based on those outcomes (on a per click basis).

2) Affiliate fees on native checkout

The other announcement was related to affiliate commerce. ChatGPT will start supporting direct native checkout of products across Shopify, Walmart and Etsy among other platforms. As part of that, OpenAI will take a 4% cut of any transaction for merchants that choose to allow native checkout (for now, just from Shopify merchants).

This method is aligned with merchants and doesn’t require adapting LLM responses, and only charges them for native checkout which is optional (they can still show up as they would in the response for no cost if native checkout is not enabled).

The importance of this is two-fold:

• Another monetization avenue: It provides another monetization path around the commerce-related queries on ChatGPT for purchases that happen directly on the platform, without having to directly sign up every brand as an advertiser. Brands are used to paying affiliates across channels and this allows them to view OpenAI as a paid channel for native checkout as they treat some other marketplaces such as Amazon, etc.

• Higher value capture of Commerce Ads budgets in future: Commerce is a very important vertical (typically 35%+ of ad revenue from Meta and Google) and so if native purchasing behavior on ChatGPT happens and improved conversion rates, that makes it easier to measure impact/outcomes and incrementally and later capture larger budgets but also more value per transaction from this universe of advertisers.

III. The OpenAI Ads Opportunity

Long-term, the Ads opportunity for OpenAI is massive and could even be a Google, Meta or TikTok scale opportunity ($50-200B). But to help contextualize the shorter term path, below is what the opportunity might look like in the short term across both avenues.

Affiliate commerce and native checkout is a relatively new phenomenon and so may take a while to get going and as you can see need aggressive adoption and annual spend to get to even a $1B business.

Intent ads in chat are relatively familiar to users, but may take OpenAI a year or two to build the required infrastructure to support self-serve advertising, measurement, the right targeting and so on to really command large budgets. As you can see below, relatively conservative assumptions based on the current scale can see this being a $2-4B business in the coming few yeas, depending on how many users stay on and continue to use the free plans. Long-term, the potential for it to be a $10B+ business even just in the US is quite clear, if they can get to high levels of ARPU.

IV. Ads for Agents

So far, we’ve been talking about ads shown to humans. I think a larger shift will happen when the interface isn’t “chat” but rather “an agent completing a task.”

If an agent is buying running shoes for you, it’s going to do the research anyway. You can’t really “persuade” it with a banner. Doing so would mean it chooses the advertisers interests over the user and users will not like that.

The way you could influence the agent is by changing the utility function in a way that is positive for the consumer.

That’s why I find Google’s direction around “Direct Offers” interesting. Direct offers allows merchants present exclusive discounts directly inside AI Mode when Google determines a shopper has high purchase intent. It’s essentially a merchant trying to convince or negotiate the agent to go with a given brand’s product to help drive the sale.

This is how they position Direct Offers:

“Advertisers should think of Direct Offers less like a standard ad and more like a salesperson negotiating a deal on your behalf. It allows you to offer special discounts to high-intent shoppers to turn browsers into buyers.”

If we stop using chat interfaces and start to delegating more decision and purchases fully to agents without being in the loop as much, my guess is ads will have to evolve on these surfaces to look like the above. Something that influences the agent, while still driving towards better outcomes for the user.

V. Opportunities for Startups

This could be a post in and of itself, but I figured it’s worth touching on what opportunities for startups this ecosystem presents1. We’ve already seen the rise of AEO/GEO companies like Profound and AirOps to support brands in AI channels organically, and the advertising/affiliate ecosystem brings forth new opportunities for founders:

• Monetization rails for AI surfaces: OpenAI will obviously build their own ad platform, but the long tail of AI services and chatbots may want to plug into an existing network that brings both advertiser demand and the SDKs for targeting, brand safety, formats, and measurement. Companies like Koah, Kontext, and Gravity are building in this category.

• An “AI-channel DSP” for brands: A buying and measurement layer that helps advertisers manage spend across AI surfaces as more of them support ads, similar to how DSPs helped coordinate spend across social platforms and CTV. The core product is cross‑channel budget allocation, creative management, and measurement and reporting built for chat outcomes (leads, conversions, purchases).

• Agentic commerce enablement for merchants: Infrastructure that helps brands support native checkout and “offers” across emerging agentic commerce standards, especially for merchants outside Shopify’s ecosystem. Think products that let merchants quickly support agentic commerce and native checkout with connectors for catalog, inventory, pricing, and payments, without forcing a full eCommerce replatform. I’ve previously written about other opportunities in agentic commerce.

V. Closing Thoughts

I’m excited to watch OpenAI’s execution over the next year around both these monetization avenues: classic intent ads and affiliate checkout. The first one is about building a real ads product, pacing supply, proving measurement, and doing it without users feeling like answers are getting “shaped” by budgets. The second one is about behavior change and whether ChatGPT actually become a place where purchases happen.

I’m also curious how quickly the rest of the ecosystem follows. Some players will copy the search-like ad unit, some like Anthropic may ignore all this completely and some will outsource monetization to companies like Koah and Kontext to get going faster. And then longer term, the agent shift could change the fundamental nature of what ads look like on these channels2. Once models start completing tasks end-to-end, the winning “ads” probably look less like banners / units to persuade humans and more like offers that make the user better off (price/discounts, guarantees).

If you’re building in and around this space, I’d love to hear from you at tanay@wing.vc

Subscribe now

Hi friends,

Two Chinese foundation model labs, Zhipu and MiniMax, went public over the past few weeks, with Zhipu becoming the first foundation model company to IPO globally. What stood out to me reading both filings is the different approaches they took to turn their models and capabilities into a real business.

Zhipu focuses on enterprise more, with a bulk of their revenue coming from on-prem enterprise deployments, while MiniMax is more product and consumer focused with multiple apps that leverage the models they build driving the majority of their revenue.

I’m going to break down both companies across product, business model, financials, and IPO reception1. Let’s dive in.

1) Product

Zhipu

Zhipu is a deployment-first model company. The core asset is the GLM model family, and the filing emphasizes a set of newer releases like GLM-4 and its variants, including lighter versions and vision-capable versions, plus newer iterations in the GLM-4.x line which are most popular for their coding abilities. Zhipu is very explicit that it delivers capability through public cloud, private cloud, and on-prem deployments, and it pairs that with professional services. In other words, the “product” is often the model deployment plus a set of services that make it usable inside a real enterprise environment. They talk about their “Model as a Service” platform, which is highlighted below.

MiniMax

MiniMax is a product-first model company that was founded in December 2021. They still talk about their core model suite as below, but what jumps off the page is the multi-modality and the consumer-facing surfaces they have built. The filing discusses their lineup of models that spans text models and models for video, speech, and music generation. But the way these are primarily delivered is via consumer apps - Hailuo AI, Talkie (and Xingye), MiniMax, and MiniMax Audio. MiniMax is trying to express model capability through native products that can acquire users and monetize directly, while also running an Open Platform that can serve developers and enterprises.

2) Business Model and Monetization

Zhipu

Zhipu monetizes primarily through enterprise solutions delivered via deployments and support.

They generated $8m in 2022, $17m in 2023, $44m in 2024, and $26m in H1 of 2025, and so went public quite early in their monetization journey, although it should be noted that many of their key model launches that put them on the map outside of China came in the back half of 2025.

Zhipu has two main offerings: An on-premise deployment where they can help either fine-tune models and host them locally within a customer’s own environment and a cloud offering where anybody, including individual developers, can use their models as a service via an API.

The on-premise deployment accounted for 84.5% of revenue. The economics of the two offerings are quite different. The on-premise deployment has only 95 customers and so much higher ACVs while the cloud offering is used by individuals and teams across thousands of enterprises. Additionally, their on-prem deployments have a gross margin of about 60%-65% depending on the quarter. Their cloud deployment is running on a gross margin of 0% to 5% depending on the quarter.

Given the on-prem model driving majority of their revenue today, it’s perhaps not surprising that their revenue is also quite concentrated. The top 5 customers represent about 40% of the revenue in H1 2025, with the largest being 11%.

MiniMax

MiniMax monetizes through a two-pronged strategy that balances AI-native consumer applications (B2C) with a B2B Open Platform for developers and enterprises. The company generated no revenue in 2022 during its initial R&D phase, but revenue grew to $3.5m in 2023, jumped to $31m in 2024, and reached $53m in the first nine months of 2025. Similar to Zhipu, MiniMax is entering the public market at an early stage of commercialisation.

While Mainland China accounted for 81% of revenue in 2023, the business has moved globally; by the first nine months of 2025, Mainland China represented only 27% of revenue, with Singapore and the United States contributing 24% and 20% respectively.

On the B2C side, the company operates four core applications that served over 212M users as of September 2025. Its largest revenue driver is the entertainment platform Talkie (Xingye), which features emotionally responsive virtual characters and recorded 20M MAUs and $19M in revenue for the 2025 period. The fastest-growing app, Hailuo AI, focuses on cinematic visual and video generation, contributing $17M. The suite is rounded out by MiniMax Audio, which provides speech and music synthesis and a specialized intelligent agent for professional tasks like coding and research. Interestingly, given the discussion about ChatGPT ads, it should be noted that their core app Talkie primarily monetizes via advertising, with Hailuo and others primarily monetizing via subscriptions and in-app purchases.

The economics of its two segments differ significantly. The AI-native products (B2C) segment accounts for ~71% of revenue but operates on thin gross margins of ~5%, having only recently recovered from deep negative margins as inference efficiency improved. In contrast, the Open Platform (B2B) segment, which represents ~29% of revenue, has ~70% gross margins. This platform provides a mix of usage-based cloud APIs and licensed deliverables that allow enterprises to deploy and operate models locally within their own internal systems.

3) Financial Profiles

It’s worth touching a bit on some of the financials in aggregate for the businesses. If you take the reported numbers given and convert them to annnualized numbers, the shape of the two businesses looks roughly as follows.

Revenue: MiniMax is a bit larger and growing faster, though Zhipu’s 2025 numbers may meaningfully accelerate in H2 ‘25 with their model launches in the back half of 2025.

• Zhipu went from $45M in 2024 revenue to ~$54M in annualized revenue

• MiniMax went from $30M in revenue to ~$71.2M in annualized revenue

Gross Margins: Zhipu has strong gross margins given it’s more enterprise / on-deployment focus.

• Zhipu’s aggregate gross margins were around 50% in H1 2025, down a bit from ~56% in 2024

• MiniMax’s gross margins were 23% in Q1-Q3 2025, up from 12% in 2024.

Losses: Both companies are burning a lot of money as one might imagine, with R&D and more specifically cloud services for model inference and training being the primary drivers of spend, at around 70% of R&D spend and R&D itself being about 65-80% of revenue for each company.

• Zhipu lost 415M in 2024, with true cash burn of ~$300M, and plans to burn ~550M in 2025

• MiniMax lost ~465M in 2024, with true cash burn of ~$260M and plans to burn ~340M in 2025

Cash: Both companies had raised over $1B prior to IPO and raised ~500M each in the IPO itself. Interestingly, at their high levels of cash burn, Zhipu in particular was down to <1 year of runway on their liquid cash and equivalents prior to the IPO.

4) IPO Reception

Both companies raised large IPO proceeds by any normal standard, and a small round by the standards of AI labs in the US that we’re routinely seeing these days.

Zhipu raised over $550M and MiniMax raised over $600M in their IPO. Both IPOs have fared really well, with Zhipu up almost 80% and MiniMax popping over 100% on its IPO day, with both pricing their IPOs in the ~$6.5B range.

Currently, MiniMax trades at a valuation of $15.6B, and Zhipu trades at a valuation of $11.7B. At these valuations, MiniMax and Zhipu trade at 2025 annualized revenue multiples of ~220.4x and ~215x respectively!

Seems like a good sign for all the various AI labs globally that perhaps want more capital or go public. In closing, here’s an infographic based on this post.

Hi friends,

I hope everyone is enjoying a restful holiday period. As we head into 2026, I’ve been jotting down a few areas in AI and technology that I’m paying closer attention to with some predictions as well. Let’s dive in!

I. Monetizing free AI surfaces

A lot of the most widely used AI products today still have an unresolved question sitting at the center: what do you do with massive free usage at scale?

Subscriptions work well for power users. Enterprise/APIs works well for teams. But neither fully addresses the reality that tens or hundreds of millions of people are using chatbots regularly, often for expensive queries.

I think 2026 is when we start seeing experiments of ads in the major chatbots such as ChatGPT, Gemini and Claude. They probably don’t take the form of ads inside answers, which would immediately undermine trust, but through more separated, carefully designed monetization. Sidebar ads. Related sponsored results. Affiliate-style ads on commerce.

OpenEvidence is reportedly already on a 9 figure run-rate from ads, and highlights that in many cases the commercial intent of these queries is extremely real and that you can’t just rely on monetizing the small fraction that subscribe to paid plans forever.

II. Proactive AI

Most AI today still waits for you to ask.

One of the trends I’m most excited about is the move toward more proactive behavior. Systems that monitor things on your behalf and surface updates, summaries, or drafts without you having to prompt them.

This doesn’t look like a general life assistant. It shows up in narrow, permissioned contexts where the AI already has the right signals. Daily briefs. Alerts when something changes. Work that starts before you ask for it.

We’re already seeing early versions of this with ChatGPT’s Pulse and Yutori’s Monitor, and I expect more products to lean into this over the next year.

III. Claude Code for knowledge work

Outside of software engineering, the dominant interface to AI is still chat. That’s fine for quick tasks. It’s a bad fit for work that is multi-step, long-running, or closer to delegation than prompting.

I think we’ll see something like the equivalent of Claude Code emerge for the typical knowledge worker in 2026. An agentic interface to manage tasks that can take a task, operate for a while, use tools, and come back with completed output rather than suggestions.

Manus is probably the closest thing in spirit today. But I think the category becomes more legible in 2026, even if it doesn’t converge into a single product. It may emerge role by role, the same way people talked about “Cursor for X,” but now in “Claude Code for X” form.

IV. Computer use as a first-class capability

Agents get much more interesting once they can operate the world as it exists today.

That means using browsers, navigating apps, filling out forms, clicking through workflows. Computer use has been an area of focus for a while and we have started to see improvements in benchmarks, but today, at least on true taking action, browser use and computer use are not still reliable enough to do most core consumer or enterprise workflows.

I think we start to see the payoff in 2026, with models that are meaningfully better at doing this reliably - operating applications, filling 10 page forms, etc.

I expect to see meaningful jump in computer use benchmarks, but more importantly in products that can make use of the capability to actually carry out work (beyond the toy book a flight example).

V. Continual learning breakthrough

One research topic that keeps coming up more in conversations is continual learning which is the capability of models to learn new tasks / memory / feedback without forgetting past ones.

Systems that can actually improve over time from experience and feedback aren’t quite there yet. Dwarkesh wrote recently about this being a meaningful blocker for more general intelligence which I agree with.

I don’t think this gets “solved” in 2026. But I do think we see a real leap from one of the frontier labs which also becomes available in the products we use and makes AI feel even more useful. An agent that gets better as it runs is fundamentally different from one that stays static forever.

VI. A blockbuster IPO

Finally, I’m watching the public markets. I think this year brings even more liquidity, in the form of all three, pseudo acquisitions, actual acquisitions and IPOs.

And maybe most importantly on the IPO front, my expectation is that we see at least one (if not more) of the $100B caliber companies file to go public in 2026. Yes, the ones that have been perma-private so far and raised boatloads in the private markets which we questioned whether they’ll ever go public (but not Stripe).

Closing Thoughts

Given how far we’ve come in AI over the last year, when reasoning models were barely getting going, 2026 promises to be another eventful and action-packed year. What are you keeping an eye on this year? As always, I’d love to hear your thoughts, and if you’re building in the first few areas discussed above, feel free to email me at tanay at wing.vc.

Subscribe now

Hi friends,

Most of us today use AI in a chat box. You type, it replies, everything happens in that one thread. That was the right starting point. It is the wrong place for a lot of the work we are now trying to hand off to agents.

Refactoring a codebase, watching the web for changes or building a briefing for tomorrow are not five second tasks. If you force those into chat and let it run for 10s of minutes, people inevitably get bored and leave.

For these tasks, The model should not be a slightly faster colleague who types back at you. It should feel more like a background process that you brief, that disappears for a while, and that hands you something useful where you actually work.

That is the shift I want to talk about: the rise of background agents, how they run, and where we already see them showing up.

What are background agents

A background agent is an agent you do not have to sit and watch. You might talk to it in chat, you might tag it in Slack, you might forward it an email, but once it understands the task, it runs elsewhere and comes back only when it has something worth your attention.

That already makes it different from a classic chatbot. In the chatbot world, every interaction is “you ask, it answers.” In the background world, the unit of interaction is a task. You brief the agent once, it may run for minutes or hours, and the result shows up as a pull request, a digest, a comment, or a notification.

On top of that, background agents can be:

• Ambient: They are always running in some sense, and responding to some changes in some input. They wake up because the world changed, not just because you typed a prompt in. A new email arrives, a web page changes, a metric moves, a calendar event appears.

• Proactive: They are allowed to tap you on the shoulder when there is something important, rather than waiting for you to ask at all. Think alerts, daily briefings, suggested decisions or actions (or at some point it may have taken these actions for you as well).

Not all background agents have to be proactive or ambient. You can have a very simple background agent that only acts when you tell it to. But once you accept that the work happens away from the chat window, it becomes natural to let agents respond to time and events as well as direct commands.

Chat is still useful. It is a great place to negotiate scope, explain edge cases, and set guardrails. It just does not have to be the only interface or the place where the heavy lifting happens.

What these agents actually look like

Broadly, background agents show up in two ways.

The first are the ones you explicitly create. Early on, that mostly happened in chat. You open something like ChatGPT and kick off a Deep Research request or a coding assistant, describe the task, and a background job spins up from that thread. Cursor or Codex style agents that take a spec, go off, and eventually return a pull request are good examples here.

What is changing is that these agents are now being created in the places where the work already lives. Instead of only briefing them in chat, you:

• Assign an issue in Linear or Jira to an agent instead of a person

• Tag an agent in a Slack or Teams thread and ask it to own that conversation

• Forward an email to an agent address which can ingest the context and get started

You are still creating a discrete job, you are just doing it from tickets, threads, and emails rather than only inside a chatbot.

The second style is more ambient and always on. Here you create the concept once, not each individual run.

That might look like:

• An agent that runs whenever new emails come into your inbox, monitoring your inbox and turning a pile of messages into a single summary and task list or drafting replies (e.g. Fyxer)

• An agent that tracks changes on competitor sites or pricing pages and hands you updates so you can react (e.g. Web Monitoring Agents such as Yutori Scout)

• An agent that watches or monitors a set of data for anomalies or changes and only surfaces something or takes action when it sees a meaningful pattern

Sometimes these run purely on events, sometimes they also run on a schedule, for example sweeping everything once a night and dropping a fresh brief in the morning (e.g. daily briefing agents that can be created on Lindy or Gumloop). Sometimes these can be derided as just cron jobs, but they also have LLMs baked in, and so are more powerful.

So you get two complementary patterns. Jobs you explicitly assign, often from the same tools you already use to assign work to humans. And agents that feel more like part of the environment, always running in the background, watching your inbox, data, or the web, and only appearing when there is something you might want to do about it.

Examples of background agents

To make this less abstract, here are three classes of background agents that already exist and feel meaningfully different from simple chatbots.

1. Coding agents as background teammates
   Cursor is probably the cleanest example. You describe a change, sometimes in chat or directly assigning it in Linear. Cursor spins up cloud agents that read and edit your repo, run tests, and open pull requests. The work might run for quite a while. When you come back, the real interface is the diff and the test results.

2. Monitoring agents for the web
   Products like Yutori’s Scouts or Exa-powered monitors allow you to create an ambient agent that continously monitors the internet for changes in things you care about. You say which companies, topics, or products you care about. The system spins up agents that continuously check relevant pages, feeds, or search results. When something meaningful changes, you get a short synthesized update. At some point, these agents may be able to take direct actions as well (buy item if price drops below certain amount, if competitor changes pricing, update our comparison page to reflect that).

   

3. Daily brief and inbox agents
   There is a growing set of tools that connect to your email, calendar, and documents and then run every night or as new emails come in. Fyxer is a great example: it monitors your inbox, and autocategorizes email and drafts responses in your voice. Another good example are various daily briefing agents that run at scheduled times overnight and review the next day’s calendar and create

You can imagine similar patterns inside companies: agents that monitor various forms of data and proactively take action.

Closing thoughts

This cycle started by putting a chat box in front of a model. The next phase is about letting agents running on their own time - it allows them to run for longer, always remain running/monitoring and even be proactive.

Background agents are the mechanism that makes long running, multi step, and recurring work feel natural. They may start from chat, but they do not have to live there. They can be user-created, or ambient and always on. They can be quiet most of the time, then proactive in short bursts when something actually matters.

My guess is that most of the real leverage from AI will come from those quiet loops: agents working in the background while we are in meetings or asleep, and then appearing, briefly, with something that is actually useful. Chat will still be there, but more as the way we brief and tune them, rather than the place where all the work happens.

If you have thoughts about this paradigm or are building various forms of these agents, I’d love to hear from you at tanay at wing.vc

Subscribe now

Hi friends,

Voltaire once wrote “This agglomeration which was called and which still calls itself the Holy Roman Empire was neither holy, nor Roman, nor an empire”

A similar confusion has crept into one of SaaS’s favorite metrics. ARR, short for annual recurring revenue, used to have a simple meaning: the predictable subscription revenue a company generates.

But lately, what some startups call ARR is neither annual, nor recurring, nor in some cases, even revenue. Let’s look at what’s going on.

1) When ARR is not annual

What’s going on

Many companies take monthly or cancel-anytime contracts (or cancel within 3 months with most customers in that period) and multiply the current month’s revenue by twelve to arrive at an ARR number, arguing that if customers keep paying next month, that’s what they’d earn in a year. The issue is that these customers haven’t committed to a year, so the figure represents potential, not commitment. We all know many of the budgets in AI today are experimental.

The other version of it is using peak spikes in a customers demand that likely won’t recur month to month and using that to annualize.

Why it matters

Annual implies time-bound commitment and predictability. When most of your base can churn in a month, the number behaves like annualized MRR, not true ARR.

2) When ARR is not recurring

What’s going on

One common stretch is project type revenue being bundled into ARR, which is especially true for service type models which have grown in the AI world that are often transactional/project-based. In some cases, those customers come back for multiple projects so it can feel recurring, but these are still discrete projects that must be re-won each time.

One of the other trickier ones is with the rise of usage-based models, whether to annualize the usage. The logic is that consumption repeats, so it must be recurring, which it may be, but in some cases there are clear spikes for certain reasons such as the first time to onboard, etc and the usage is likely to not recur in the future.

Finally, some companies are baking pilot revenue into ARR by including the revenue from the pilots themselves or sometimes the assumed size of the contract when it is signed (gulp).

Why it matters

Recurring means repeatable by contract or strong customer habit. When project work or speculative expansion is blended into ARR, it makes the metric lose meaning. Distinguishing recurring from repeatable or re-occuring helps shed better light on the business.

3) When ARR is not Revenue

What’s going on

Perhaps the most creative accounting happens here. Some companies are treating gross merchandise volume (GMV) as revenue, even though the company only earns a small take rate.

Others are annualizing pilot conversions, counting the potential revenue “if all goes well” as if it already exists, and in some rare cases, companies are including potential expansion figures which haven’t happened yet into ARR (again, gulp!).

Why it matters

Revenue represents money that has actually been earned by the company and is supported by delivery or performance. Counting unearned expansions or total GMV overstates reality, since those dollars may never materialize or do not belong to the company in the first place.

Closing thoughts

The pressure to show eye-popping growth is real. Screenshots of “$0 to $5M ARR in X months” and soundbites of “We only invest in $0 to $2M ARR in 10 days” travel fast, and it’s tempting to keep up.

If your business relies on short-term contracts, usage, or pilots, that’s fine. Many companies in the AI world are the same. Just label it honestly as annualized run-rate revenue or annualized MRR and remember that it is one of the aspects of the business to focus on and improve over time. Revenue quality ultimately decides who endures. Retention and gross margin matter more than vanity milestones.

Subscribe now

Hi friends,

It’s OpenAI’s world and we’re just living in it. Over the past few months, news tying a company to OpenAI, whether official partnerships or announcements or just chatter has moved stock prices of large companies across chips, cloud, commerce, and beyond.

OpenAI seems to be having a kingmaker effect in the public markets unlike anything we’ve seen before. Great for partners and those who can be aligned with their direction, terrible for those in the firing line.

Note that I won’t focus on the accounting treatment or circular nature of these deals, but instead on just how broad and impactful announcements with OpenAI are proving in the stock markets across the board.

Compute & Cloud

NVIDIA
On September 22, 2025, NVIDIA and OpenAI outlined a plan to deploy at least 10 gigawatts of NVIDIA systems for OpenAI’s next wave of infrastructure, alongside press about a large prospective NVIDIA investment linked to deployments. NVIDIA Stock surged 5% on the news, adding hundreds of billions in market cap.

AMD
On October 6, 2025, AMD announced a multi year agreement with OpenAI to stand up roughly 6 gigawatts of capacity, beginning with 1 GW targeted for 2H26, plus a warrant package for up to 160 million AMD shares for OpenAI. AMD rose almost 30% that day adding ~$100B in market cap.

Broadcom
On September 13, 2025, reports tied OpenAI to a roughly ten billion dollar custom accelerator program at Broadcom, and AVGO pushed to fresh highs into late September. On October 13, 2025, Broadcom and OpenAI formally announced a 10 gigawatt custom accelerator collaboration, and the stock surged nearly 10%, adding ~$100B in market cap.

Oracle
On September 11, 2025, coverage of a very large multi year OpenAI capacity agreement in the ~$300B range resulted in Oracle logging its biggest one day gain in decades, up over 35% adding over ~$300B in market cap.

Commerce

OpenAI’s agentic commerce announcements also saw a number of beneficiaries that have been early with them to partner to support agentic checkout in ChatGPT.

Etsy
On September 30, 2025, OpenAI launched Instant Checkout inside ChatGPT with Etsy as the first live surface. Etsy jumped about 16% on the news, adding ~$1B to its market cap.

Shopify
Around the same time as above, OpenAI said the same Instant Checkout surface would extend to Shopify merchants soon, implying better discovery and potential for higher sales and purchase volume over time for Shopify merchants. Shopify gained ~6% on the news, adding ~$10B in market cap.

Walmart
Earlier today on October 14, 2025, Walmart announced a partnership with OpenAI to enable shopping through ChatGPT using Instant Checkout. Walmart is up over 5% adding ~$40B in market cap.

Negative Moves

Moves aren’t always positive however as some of the examples below show

Reddit
On October 1, 2025, third party data showed ChatGPT referencing Reddit less than it had in mid September. Reddit fell roughly 10 percent on the day as the market repriced how it expected Reddit user growth, referral traffic and data licensing revenue to fair over time with decreased ChatGPT visibility, losing almost ~$1B in market cap.

Docusign
On October 1, 2025, DocuSign plunged ~12% immediately after OpenAI revealed its internal contracting agent, DocuGPT, despite it just being a demo about an internal tool. Investors got spooked about the threat to SaaS products like Docusign with LLM advancements. The drop translated into a ~$1.5B decrease in market cap.

Closing thoughts

OpenAI has become a market scale allocator of demand. One announcement of some relationship with OpenAI can send a stock soaring. One news article about OpenAI encroaching in some way can send a stock tanking. A private company’s roadmap has become one of the major inputs for the public markets writ large. The tweet below encapsulates it well.

Subscribe now

Hi friends,

ChatGPT released a really interesting paper last week that went deep into how it's being used by its over 700 million monthly active users over the last year.

About 73% of its usage is now for non-work tasks, with 23% for works tasks. But perhaps most interesting was the graphic below which broke down how people are using ChatGPT for different use cases and different “jobs to be done”. It clearly shows that ChatGPT has become an all-in-one assistant across personal and work tasks and multiple jobs to be done.

Some of these jobs to be done required others websites or tools to serve the same need, and some of these jobs to be done are things where there wasn’t a good parallel prior to the advancements in LLMs. Let’s discuss further.

The bundle that is ChatGPT

“There are only two ways to make money in business. You can bundle, or you can unbundle.” — Jim Barksdale.

Today, it’s clear that ChatGPT is basically used as an all-in-one tool that bundles and absorbs many jobs that used to live in separate products.

Some of the top use cases of ChatGPT per the chart are:

• Practical Guidance (28.3%) which includes brainstorming ideas, getting how to type help on aspects of life, fitness, self care as well as Tutoring or teaching. These are things which people relied on websites such as YouTube, Reddit, etc before for the former, and YouTube, KhanAcademy and Chegg for the latter, but were not personalized or adaptable to any question.

• Writing and editing (28.1%) including writing drafts of posts, emails, etc, editing and summarizing existing text. This is something which may have been done by things like spellcheck, Grammarly, and parts of these (first drafts from scratch) really weren’t possible or accessible before.

• Seeking Information (21.3%) which is very clearly a form of search and information retrieval which people relied on Google, specific vertical blogs and forums for.

Below is a version of the chart showing what various tools and products ChatGPT can “substitute” across the various use cases by fulfilling the job to be done itself.

Bundling wins here for simple reasons. One interface replaces a pile of tabs/apps, which cuts switching costs and makes the next step instant. A single memory carries context across jobs, so the draft you wrote informs the summary you ask for, which informs the email you send. Coverage matters too. ChatGPT is good enough across many jobs in the same session, which is how real work and life actually happen and so if you aren’t a frequent user of a use case, you never need to bother finding the best product for it. And then ofcourse, speed improves because the model does the next action (giving you the answer) rather than handing you links.

Unlocking Net New Behavior

But it is not just that ChatGPT can perform many jobs other tools could do. It enables jobs that are bigger and more fluid because of how LLMs work. The paper frames usage as Asking, Doing, and Expressing. Asking is about getting information or advice. Doing is asking the model to produce an output you can plug into a process. Expressing is sharing views or feelings.

Across the sample, about 49% of messages are Asking, 40% are Doing, and 11% are Expressing. For work messages, about 56% are Doing, and most of those are writing tasks. By late June 2025 the split was 51.6% Asking, 34.6% Doing, and 13.8% Expressing. The chart below shows the split across these.

Why does this matter? Doing is the net-new capability at consumer scale. The assistant does not only answer questions (which in and itself is better than the past form of getting answers manually). It produces artifacts and chains steps with context that carries across turns. You ask for a plan, then ask for the spreadsheet version, then ask for a manager-ready summary, all in one thread. You describe constraints and get a finished draft that fits your situation rather than a generic page of links.

This is what lifts the ceiling on old jobs to be done like tutoring, cooking, or data pokes, and it also creates new habits that did not exist for most people before.

What comes next

Bundles create demand by collapsing steps and reducing friction. Specialists then peel off power users who do one job all the time and need depth, speed, and tight integrations. That is when unbundling shows up. You can already see the signs of it.

For those that code a lot, specialized products such as Cursor and Windsurf that leverage the capabilities underlying ChatGPT (i.e., the models) are better. For those that create a lot of images, using Midjourney or Krea may be better. For those that are writing for work, Notion that has the context of your related knowledgebase may be better.

If you do one of these tasks heavily in your job, a focused tool will often beat the general assistant that is ChatGPT. That idea of the unbundling that is to follow and where we are likely to see it next is what I will cover next week.

Subscribe now

Hi friends,

A lot is happening in AI browser land, so let’s talk about it. Just this past week, Atlassian agreed to buy The Browser Company, the makers of Arc and Dia for $610M in cash and a federal judge declined to force Google to sell Chrome, which keeps the incumbent’s distribution power intact. Perplexity continues to roll out Comet more broadly. OpenAI is widely reported to be prepping a browser of its own.

Let’s jump in and discuss:

• Why the browser matters

• Who the major AI browser contenders are

Why the browser matters for AI

I expect the browser to be an interesting battleground going forward for the following reasons.

1. Distribution surface: Default slots in browsers are worth billions. Google paid Apple about $20B in 2022 to be Safari’s default search on iOS, and spent $26.3B in 2021 across devices and platforms to secure default status. A recent ruling lets Google keep paying partners for defaults, which preserves that distribution engine. Mozilla’s finances also show how powerful this channel is, with a large majority of revenue tied to its Google search deal. Search was the original beneficiary of these defaults, and chat assistants are now stepping into the same front door.

2. Place where agents happen: The browser already sits where work happens, so it is a natural potential control plane for agents. You can grant page context, watch steps, and take over when needed. That is why new products position the browser as an agent desk. Perplexity’s Comet launched as an agentic browser that researches and acts from the page. The Browser Company’s Dia makes it easy to use AI skills and assistants using the context of a webpage. Together, these moves show agents gravitating to the browser surface first, particularly as human in the loop happens. The fact that the browser also holds identity, cookies and app state across various apps and tools people use also further helps further this idea.

The AI Browser Wars Contenders

While still relatively early, and I expect all the key browsers such as Firefox, Edge, Safari, Brave to add AI features, I figured it’s worth touching on the dominant browser today, as well as the key AI-native browsers.

Google Chrome

Chrome is the dominant browser today with almost 70% market share, and as the recent Google antritrust ruling asserted does not need to be divested by Google.

We’ve already seen a few AI features added in Chrome over the last year or so, and as the company continues its focus on AI, I expect to see more. So far, they’ve shipped things like features to organize tabs and help people write and summarize content on a page. Over time, there may be page-grounded actions across multiple tabs.

Google’s goal with Chrome is likely to funnel and leverage Chrome’s distribution to continue to maintain Google and its AI products as the defaults for most users as well as to continue to collect more data around users that can be helpful to improve and personalize Google’s products.

Given that many AI features can be quite expensive, particularly as you move into agentic features, I do wonder if Chrome's massive scale ends up being a limiter to the amount they can push with AI functionality in the browser in the absence of requiring subscriptions to Gemini or similar.

Dia

Dia was built by the team over at The Browser Company of New York, who also built Arc, and was acquired by The Issue Tracking Company of Australia (Atlassian) for about $610 million in cash last week. The Dia browser itself only launched in June of 2025, barely 3 months ago, and was the team’s take on an AI-native browser, which combined their learnings from Arc with what’s possible with AI.

On AI features, Dia already lets you chat with pages and YouTube videos, run reusable Skills, use all your browsing history as context and personalization for chats and more. The vision going forward is to make the browser more useful for work across the myriad of tools people use, including of course Atlassian’s own tools.

“Today’s browsers weren’t built for work; they were built for browsing. This deal is a bold step forward in reimagining the browser for knowledge work in the AI era. Together, we’ll create an AI-powered browser optimized for the many SaaS applications living in tabs – one that knowledge workers will love to use every day” - Mike Cannon-Brookes, Atlassian’s CEO and co-founder.

Going forward, I expect to see more agentic features that allow users to prompt taking actions directly from webpages or their browser, particularly in a work context.

If you don’t yet receive Tanay's newsletter in your email inbox, please join the 10,000+ subscribers who do:

Subscribe now

Perplexity’s Comet

Perplexity launched Comet on July 9, 2025, positioning it as an agentic browser that researches, summarizes, and then acts on the page. It is a natural extension of Perplexity’s retrieval strength, which makes answers feel fast and grounded. Distribution has ramped quickly through a broad public launch and partner promotions, and Comet has the most advanced set of agentic features and functionality of browsers today.

On AI features, Comet blends page awareness with suggested next actions and can already reach into basics like email and calendar when you grant permission. The promise is that you can ask a question, follow links, and complete simple steps without context resets. What comes next is breadth. Shopping, scheduling, and forms are obvious surfaces, along with longer task chains that need retries and recovery without losing the thread, which will take models to improve as well.

From Perplexity’s point of view, it’s likely they view the browser similar to Google Chrome - as a way of increasing Perplexity subscription adoption as well as Perplexity usage by being the default for Comet users. Interestingly, I know of some people subscribing to Perplexity’s max plan specifically for the browser itself, given how useful the still early agentic functionality can be. It will be interesting to see if it remains a distribution channel to core Perplexity or if it becomes an independent product of its own that people pay for.

OpenAI

OpenAI’s browser is not out yet, but multiple reports say it is in the works, with the internal name Aura surfacing in code per a piece in June of this year. The expected shape is a tight coupling between ChatGPT for intent capture and Operator/ChatGPT agent for tooling/an agent that can act with clear prompts and consent.

Prediction markets currently estimate a 70% chance we see the browser this year. Whenever it does come, it will be interesting to see whether given ChatGPT’s runaway lead as the AI assistant of choice for most users with over 700M MAU, they can translate that into browser downloads and usage as a way to continue to deepen their relationship with their users and increase engagement with their core product.

Closing Thoughts

The browser matters because distribution still sets the table and because the tab is a natural place for agents to work alongside us. Chrome remains the default gravity well and is layering in AI features inside the browser. Dia feels like the most AI-native take on a work browser and now has Atlassian’s distribution and product surface to grow into real workflows. Comet is the most agentic today for research and quick tasks, and it is pushing the action model forward in public. OpenAI is the wildcard. If it ships a browser that cleanly hands off from chat to page actions, it could convert it’s large ChatGPT user base into a new OS for personal or professional work.

One final thought on share and switching. Chrome’s scale means even small shifts create room for meaningful new businesses. AI feels like a large enough wedge to change habits and perhaps encourage many people to consider switching their browser, which they likely haven’t changed in a decade.

Subscribe now

Hi friends,

I hope everyone had a great labor day weekend. There has been a lot of discussion about gross margins in AI land recently, so I figured I would chime in with some thoughts. First, I’ll give a quick lay of the land of where we’re at gross margins wise across the stack. Then, a short list of considerations on the gross margin debate for builders across the stack, particularly at the application layer.

Lay of the Land on Gross Margins

• Chips: Nvidia still captures premium economics. Excluding one-off charges, Gross margins for chips hover in the 70% range consistently across quarters.

• Cloud: Hyperscalers do not break out product gross margin cleanly for AI, so operating margin and corporate gross margin disclosures are the best signals. AWS reported 33% percent operating margin for the quarter and 36.7% on a trailing twelve month basis. Microsoft reported Microsoft Cloud gross margin percentage at 69 percent, with a note that AI infrastructure scaling is pressuring the percentage. Google Cloud delivered roughly a ~21% operating margin. Net, the platforms are monetizing AI demand even while they fund the buildout. However, AI is pressuring gross margins, as called out by Microsoft, and some of my estimates have the gross margins in the 50-55% range, though they could be lower for the other clouds.

• Models: Outside estimates peg OpenAI’s gross margin roughly around ~50% while Anthropic’s numbers per the Information also hover around 60%. It should be noted that their are a number of business lines such as the API and the consumer product which make up those numbers, and each looks pretty different. It should also be noted that training costs do not fall under these gross margins.

• Applications. This is where the spread is widest. Bessemer’s 2025 dataset shows fast-ramping AI “Supernovas” averaging about 25% gross margin early on, while steadier “Shooting Stars” trend closer to 60%. They also note that many of the AI Supernovas have negative gross margins, something we don’t tend to see often in software.

  

  Additionally, the way companies are calculating their own gross margins may be different from each other, as the chart below from The Information notes. Some companies include the cost of free users within it, while others do not.

A Few Considerations on Gross Margins

The Application layer is the area which has seen the widest dispersion in gross margins, and I’ll be touching on a few considerations for builders to keep in mind. Coding apps are probably where the Gross margin debates come up most heavily, so I’ll use those as examples throughout.

1) Inference costs are falling, but the benefits accrue unevenly
There’s been a lot of talk about inference costs falling 80-90%+ per year. And while that’s true for a given model, the price of top-end models has stayed steady or even gone up as Ethan noted in the excellent analysis and piece below.

The key question is simple. Does your use case require the top model on every request, or do you only need to meet a quality bar. If you can meet a bar, routing lets you send most traffic to cheaper models and burst to the frontier when needed. If your users demand the best every time, you need pricing that mirrors usage, or margins will compress as quality expectations rise.

2) Control and workflow depth decide unit economics.
Tied to the above, when users insist on the highest performing model, a companies COGS rides someone else’s price card, the model provider. Coding assistants feel this more than almost any other category because quality is visible on every keystroke. In fixed workflows like document processing or voice IVR, vendors own the acceptance criteria, which lets them route to cheaper models by default and escalate only on hard cases. Over time, that gives them the opportunity to push app-layer margins toward SaaS-like territory.

Additionally, workflow depth matters as well. Many AI companies are derided as “wrappers” and given as the reason for low gross margins. Classic SaaS was, in practice, a wrapper over databases and public cloud too, yet it sustained 70-80% gross margins. The path is to get deeper and take control so buyers see more than “access to a model.” Control which is discussed above is one area, but good old workflow depth continues to be another. Collaboration, versioning, audit, analytics, governance, and integrations into the rest of the stack as well as going deeper into the workflows continue to be tried and tested ways to grow ACVs without large increases in AI COGS.

If you don’t yet receive Tanay's newsletter in your email inbox, please join the 10,000+ subscribers who do:

Subscribe now

3) Do not live or die on tokens alone

As hinted at above, there are more ways to make money than marking up tokens. Many of the companies, even in areas like coding and vibecoding where a lot of the value comes from the model itself, are seeking other opportunities to monetise which will not be tied to AI inference costs and so grow aggregate Gross Margins. A few of these examples include

• Hosting and deploy. Bolt and Replit monetize runtime, bandwidth, storage, domains, and private deployments once projects live inside their environments. That raises ARPU and decouples margin from token pricing.

• Marketplaces and services. Replit’s Bounties take a ten percent fee from posters, which is a clean non-inference take rate.

• Advertising / Affiliate. OpenAI has been piloting checkout inside ChatGPT with Shopify, which would create commission revenue on the free tier. Perplexity has been testing sponsored follow-up questions. It’s likely only a matter of time when some forms of ads finds its way to most consumer chatbots.

4) Pricing models still need work and iteration

Most companies started with simple per-seat plans and ran into power-user cost spikes. The shift toward other pricing models that make costs visible and predictable. Examples include seat plus pooled credits, usage with an included allowance and clear pass-through, token bundles with rollover, and BYOK for heavy cohorts. Replit has iterated across these ideas, pairing seats with credits, layering pay-as-you-go, and adjusting tiers as usage patterns change. Anthropic noted that on their original pricing structure of Claude Code, they were losing tens of thousands of dollars per month on a user on a $200 plan.

Pricing models will continue to evolve and be a lever where companies can play around with to make gross margins improve, particularly if hurt by a small subset of extreme power users.

The chart below shows Replit’s gross margins change drastically over time while revenue continued to grow as they adapted pricing models.

5) Net Margins Matter

There’s been a lot of emphasis in recent weeks on the gross margin profiles of these companies. But what’s not discussed as much is that ultimately it’s a trade-off these companies are making where they trade-off gross margins for faster growth and lower costs in other areas.

You can trade some gross margin for distribution if product-led growth is truly working which results in faster revenue growth and leads to lower sales and marketing (and perhaps even G&A) as a percentage of revenue. The Supernova pattern that Bessemer highlighted is exactly that. Very high ARR per employee and thin early gross margins that thicken over time with the above additions, but much lower S&M, G&A as a percentage of revenue than their SaaS counterparts given the fast scale and the bottoms up motion truly working.

The goal is not a perfect gross margin in isolation. The target is a healthy net margin profile as cohorts mature.

Subscribe now

Hi friends,

The three hyperscalers just reported earnings, and the cloud wars are heating up. This quarter was especially important because it shows how AI is reshaping not just growth but share dynamics. Let’s break down the numbers, see who captured the most new dollars, and then look at what management is saying about demand.

Note that these numbers are estimates, given the companies don’t report apples-to-apples comparisons on run-rate, etc and break out their cloud divisions in the same manner.

1. Where they stand today

• AWS is still the largest by a wide margin, generating about $30.9 billion in cloud revenue last quarter. That translates into an annualized run rate of roughly $124B. Growth came in at 17.5% y/y.

• Microsoft Azure continues to close the gap. While Microsoft doesn’t break out Azure run-rate directly, they did reveal that they did 75B in revenue over the llast year which we we can use to estimate their last quarter revenue of around $21B, putting its exit run rate near $84B. That’s up 39% y/y, making it the fastest grower of the three.

• Google Cloud reported $13.6 billion in revenue, which annualizes to more than $54.4 billion in run rate. Growth was 32% y/y. It’s important to note, though, that Google’s number includes Workspace (G Suite), which inflates the headline compared to AWS and Azure’s pure cloud businesses. Even adjusting for that, Google is still growing briskly, but from a smaller base.

The numbers are summarized below.

2. Net new run-rate and share capture

It’s one thing to look at total revenue, but the perhaps more interesting lens is to focus on the net new run rate added over the past twelve months. That’s where you see who is truly winning incremental share.

When you add it all up, the three of them added about ~55B in incremental annualized runrate over the last twelve months.

• Azure captured the largest slice of new dollars, adding about $23.6B in run rate. That’s roughly 43% of all the incremental spend in the market.

• AWS followed with around $18.4B, which equates to 34 percent of the net new pool. Despite its massive base, it’s capturing a smaller share of growth than it used to.

• Google Cloud pulled in about $13B, good for 24 percent of the new dollars. It’s still the smallest of the three, but the pace shows it’s carving out a meaningful seat at the table.

On a relative basis, the share of incremental dollars between the three only looks like below, with Azure taking the crown. Note that since Google Cloud includes Gsuite, it’s actual share is lower and the other’s is higher.

If you don’t yet receive Tanay's newsletter in your email inbox, please join the 10,000+ subscribers who do:

Subscribe now

3. What management is saying about AI demand

AWS notes that demand still outstrips capacity

Andy Jassy: “On the supply constraints as it relates to AWS and what we see there, -- as I mentioned, we have more demand than we have capacity at this point. And I think that -- and you see some of the constraints and they kind of exist in multiple places, the single biggest constraint power. But you also see constraints off and on with chips”

… and will continue to invest heavily in capex

“Brian Olsavsky( CFO): “Now turning to our cash CapEx, which was $31.4 billion in Q2. We expect that to be representative of the quarterly rate for the back half of this year, putting us on track for ~$118 billion full-year CapEx.”

Microsoft notes their AI leadership

Satya Nadella: “We continue to lead the AI infrastructure wave and took share every quarter this year.

… and that they also are capacity constrained

“Even as we continue bringing more data center capacity online, we currently expect to remain capacity-constrained through the first half of our fiscal year.”

Google noted their growth in large deal sizes

Sundar Pichai: “We see strong customer demand, driven by our product differentiation and our comprehensive AI product portfolio. Stats show this: One, the number of deals over $250 million, doubling year-over-year; two, in the first half of 2025 we signed the same number of deals over $1 billion that we did in all of 2024; three, the number of new GCP customers increased by nearly 28% quarter-over-quarter.”

…. and their differentiated approach of vertical integration

“We are seeing significant demand for our comprehensive AI product portfolio. Of course, this is all possible because of the long-term investments we have made in our differentiated full stack approach … which spans from AI infrastructure to world-class research, models and tooling and our products and platforms that brings AI to people all over the world”

4. Key takeaways

In closing, what do we take away from all this? A few things stand out to me:

A/ AI is the engine with capacity the bottleneck

All three CEOs hammered on AI demand as the main driver of cloud growth, but each also admitted they’re constrained by infrastructure buildouts. Backlogs continue to grow, with Google’s backlog hitting over $100B.

B/ Azure is the clear share gainer

Capturing the largest slice of new dollars shows Microsoft’s strength in AI. I wonder how much of the share gain is down to the OpenAI relationship and indeed directly serving most of the OpenAI demand.

C/ Capital Expenditure, primarily on AI continues to be significant

Lastly, it continues to be evident that there’s no slowing down on the investment front from any of the hyperscalers. Google raised its 2025 guidance to about $85B, up from $75B, to keep pace with AI-driven demand. Amazon is on track to spend roughly $118B, fueled by investment in chips, datacenters, and power. And Microsoft expects to spend over $80B, with some quarters alone topping $30B.

Subscribe now

Hi friends,

A few months ago, I wrote about the rise of the agentic workforce, focusing on the agentic coworkers popping up across the key functional roles. Today, I’m breaking down the rise of verticalized agentic coworkers. These are AI agentic coworkers built for specific roles within industry verticals.

What do verticalized agentic coworkers look like?

• Modeled on real jobs: Each agentic coworkers is designed to take a subset of core tasks of a specific, high-headcount role. Think insurance claims adjuster, nurse, or compliance officer. The focus is on work that’s high-volume, repetitive and mostly measurable and/or wasn’t getting done because of lack of time (or in job roles with labour shortages).

• Integrated into existing workflows: These coworkers plug directly into the software and tools already used by the team. They read and write data, process transactions, and interact with other digital systems.

• Task ownership: Unlike generic assistants, these coworkers actually complete work and focus on a subset of tasks that humans in the role perform, ideally taking it off the plate of the human (minus human verification).

• Usage-based or Outcome-based pricing: Pricing is usually tied to usage or results (claims processed, reports generated), not just a monthly seat fee.

• Human oversight/manager: Most agents still operate with oversight. Humans review outputs, handle exceptions, and step in when needed. The goal is to expand autonomy as trust and performance improve and expand the set of tasks they can perform over time.

Where are we seeing verticalized agentic coworkers?

While still in the early innings, we’re starting to see the first wave of agentic coworkers across most key verticals such as finance, healthcare, legal, insurance, retail, logistics, and many more. A market map is presented below of the key AI-native companies that offer at least some agentic functionality in various vertical roles.

Some of they may fit across multiple roles, but I’ve tried my best to bucket them based on the core role they serve.

Common Wedges for Agentic Coworkers

Most verticalized agentic coworkers wedge into the workflow by starting with a repeatable, high-volume task where cost of errors is relatively small. The most common entry points we’re seeing across verticals are:

• Document processing and parsing: Extracting, classifying, and acting on information from contracts, invoices, intake document. This is often the first workflow AI agents can automate end-to-end.

• Voice agents for intake, triage, sales, or customer experience: Voice-based agents are making it possible to automate high-volume, front-line interactions. Many AI coworkers handle inbound calls or chats for scheduling, triage, sales qualification, or customer support.

• Content creation: Drafting proposals, memos, emails, or reports. Agents trained on industry-specific templates can generate high-quality output that matches the expectations of the role.

• Search and information retrieval focused workflows: Surfacing relevant policies, research, or case history in response to specific queries, coupled with forms of content generation. Strong retrieval capabilities are often the backbone of agents in legal, finance, and healthcare, that start of looking like assistants and over time can draft forms, memos, etc

I’ve written this post which goes deeper into these wedges.

Closing Thoughts

Most of the AI coworkers do a fraction of the tasks of a human in their role today. But as models and agentic systems get better over time, they will be able to do more and more tasks and go from junior interns to junior employees to beyond that. It’ll be interesting to see how these companies evolve over time. A few different combination of approaches companies may take:

• Do more work within the role: Eat up more and more of the work (different tasks) and increase their value and pricing power over time

• Build adjacent coworkers: Build adjacent coworkers for additional roles within the industry to capture more spend within customer base.

• Build AI-native systems of action: Leverage the data they’re collecting through doing work to build an AI-native system of record and action within the industry they operate in.

Subscribe now

Hi friends,

Figma just dropped its S-1, and it's a fascinating one. The company, famously at the center of then abandoned $20 billion acquisition by Adobe, has emerged stronger and more compelling than ever. There’s a lot to unpack, from its patient origins to becoming an elite, multi-product platform nearing a billion dollars in ARR. And it's navigating all this at a critical moment in AI.

Here’s how I’m breaking it down today:

• Origin Story and Core Product

• Product Vision and Multi-Product Evolution

• Business Model

• Financial Profile and Efficiency

• The Go-to-Market Engine

• The AI Wildcard

Let's jump in.

Origin Story and Core Product

Dylan Field and Evan Wallace founded Figma in 2012, aiming to create a browser-based alternative to traditional design tools like Photoshop. Unlike many startups today that chase immediate revenue, Figma spent over four years focused exclusively on product development. The company launched publicly in 2016 and only began generating revenue a year later.

Relative to some of the $0 to $10M ARR in a year from founding type of companies of today, Figma was a slow burn and took a long time to get the product right. Figma was founded in 2012, raised a seed and Series A pre-launch, launched in 2016 and started to monetise in 2017.

Figma’s founders and early investors deserve significant credit for their patience and foresight, prioritizing getting an MVP with a high bar and focusing on the leading indicator of user engagement over early monetization. The chart illustrates this idea that it took a while to get escape velocity, and many of the early rounds were pre-product and very early revenue. Vision and then product engagement were the leading indicators.

The foundational thesis was ambitious: browser-native, real-time collaborative design powered by the enabling tech of WebGL for desktop-level performance. This collaborative approach became central to Figma's value, enabling product teams across roles, not just designers, to work seamlessly within a single tool.

Today, two-thirds of Figma’s 13 million monthly active users are non-designers, highlighting how broadly its vision has resonated.

Product Vision and Multi-Product Evolution

Figma has evolved far beyond its original design tool, becoming an essential platform for entire product development cycles. Its products span all stages of the cycle as below:

• Ideate and Align:

  • FigJam (launched April 2021): An online whiteboard for ideation, brainstorming, and rapid communication, designed for users of all skill levels with features like timers, voting, diagramming, and AI-generated templates.

  • Figma Slides (launched June 2024): A collaborative presentation tool that combines high-fidelity design with real-time multiplayer editing, enabling teams to create engaging presentations, embed prototypes, and seamlessly transition from brainstorming to slides.

  Visualize

  • Figma Design (launched September 2016): The core browser-based UI design tool, offering high-performance vector editing, prototyping, and a robust design system to visualize ideas at any fidelity and enable collaborative design at scale.

  • Figma Draw (launched May 2025): A dedicated space for advanced vector editing within Figma, tailored for creating detailed iconography and product illustrations that require fine control.

  Build and Handoff:

  • Dev Mode (launched June 2023): A specialized workspace for developers to inspect designs, access specs, generate platform-specific code, and connect design systems to codebases, streamlining the design-to-development handoff.

  Ship:

  • Figma Sites (launched May 2025): Lets users design and directly publish responsive websites to the web with custom URLs, supporting features like breakpoints, code layers, and a built-in CMS for easy content updates.

  • Figma Buzz (launched May 2025): A tool for creating and scaling marketing assets such as social media graphics and digital ads, offering templates, bulk asset generation, and AI-powered editing for users of any skill level.

The evolution of this over time is interesting as well and show below, which highlights the signaling an increasing acceleration in the multi-product journey around their vision above particularly since 2023.

The core product remains the Figma Design tool, which holds a somewhat monopolistic position in the core product design market, with as high as 80% according to some surveys. But Figma has leveraged that core product to expand into adjacencies as part of the product development process well.

Per the most recent quarter, 76% of customers now use multiple Figma products, confirming early success in cross-selling and platform expansion.

Business Model

Figma operates a straightforward SaaS subscription model, charging customers based on seat-based licenses. Pricing tiers range from free plans (individual users and small teams) to Professional, Organization, and Enterprise plans, scaling with additional features, admin controls, and collaboration capabilities.

There are different types of seats depending on the use type and the suite of products they need, but range from $5/mo to $90/mo depending on the tier and type of seat.

Figma has 13M MAUs, and 450K paying customers. Customers have multiple seats and the paid MAU isn’t disclosed, but it’s likely that the vast majority of the 13M MAUs are on a free plan.

Historically pricing has been seat-based and relatively straightforward, allowing easy adoption and predictable revenue growth. However, with the integration of AI-driven products like Figma Make there is potential for the business model to evolve potentially introducing usage-based or value-based pricing models to reflect enhanced productivity from AI tools.

If you don’t yet receive Tanay's newsletter in your email inbox, please join the 10,000+ subscribers who do:

Subscribe now

Financial Profile and Efficiency

Figma’s financial profile is impressive, characterized by strong revenue growth at an impressive scale nearing a billion in ARR, coupled with profitability and an efficient business with strong retention

A few key metrics:

• Revenue and growth: Figma is at $912M in ARR growing 46% y/y

• Gross margins have consistently been around ~88-91% range

• Operating Margins are now at 16% based on Q1’25, non-GAAP operating margins slightly higher around ~18%

• Free Cash Flow Margins (adjusted) are around 27%

• Rule of X: Figma’s Rule of X is in the ~62% range (using Revenue Growth + GAAP operating margin) or if using Revenue Growth + FCF (with adjustments) would be at 73%. Whichever way you look at it, this would put it very close to the top of the public software universe.

• Net Dollar Retention Rate of 132% (for users paying >$10K) and Gross Dollar Retention Rate of 96% which is excellent although defined slightly oddly1.

Revenue breakdown

Figma has 450k paying customers, and $912M in ARR. That corresponds to an aggregate ACV of ~2K. However, the distribution is quite wide, particularly given the PLG motion and plan, and so a better breakdown is as below. About 2/3 of the revenue comes from customers paying >10K/yr.

78% of the Fortune 2000 and 95% of the Fortune 500 use Figma highlighting their reach. At the same time, only 28% of the Fortune 2000 spend >100K on Figma highlighting the potential ahead.

Cost Structure

As in the chart below, Figma has become more efficient over the last few years, particularly on G&A as well as a little bit on the sales and marketing side. Figma now spends more on R&D than it does on sales and marketing and on a GAAP basis has +16% operating margins.

Burn over Lifetime

Figma has been an extremely efficient business over its history. It currently has ~$1.5B in cash on hand as a business. In its aggregate over the course of being private, it raised $333M in primary capital until 2021 (and another $400M in 2024 out of which arguably only $60M was truly primary). It obviously received $1B from Adobe as part of break-up fee, of which ~$297M was paid in taxes/transaction costs, meaning $703M net cash received.

So Figma raised ~$393M in true primary capital and received another ~$703M net received from Adobe which is $1096M, but has $1.5B in cash on hand, and so has actually generated almost $400M in free cash flow as a business over its lifetime, net of the additional from Adobe.2 Very few venture-backed businesses go public burning a small amount, let alone a negative burn of that magnitude!

Magic Number

Another metric to highlight Figma’s efficiency is the magic number which has tended to hover in the 0.75-1.25 range which is quite impressive given the scale of the business. I’ll touch on it next, but part of the reason for this is the PLG engine the business has been able to build over time

Go-to-Market Engine

Figma is one of the poster children of the Product-Led Growth (PLG) movement. With around 450,000 paying customers and 13 million monthly active users, the company's growth has been driven primarily by bottoms up adoption and word-of-mouth growth within and across companies.

About 70% of now Organization and Enterprise plan customers initially adopted Figma through the Professional plan in that at least one user was on that plan, illustrating their effective land-and-expand strategy fueled by their product-led growth motion.

A few points to note which has enabled their growth motion and become part of the PLG playbook of sorts:

• Free Starter Plan: Figma offers a free Starter plan designed for personal projects and new users. This plan provides an easy entry point for individuals to try Figma’s design and collaboration tools without any upfront cost

• Easy self-serve onboarding: Users can sign up, onboard, and start using Figma directly through the website, without needing to interact with sales which now feels quaint and commonplace, but wasn’t the norm at the time, particularly for this kind of software.

• Built-in virality with URL sharing: Figma’s real-time, browser-based collaboration means any file can be shared instantly via a URL. This makes it effortless for users to invite teammates, clients, or stakeholders, driving organic adoption within organizations

• Figma for Education Program: Figma offers the Professional plan free for educators and students and the Organization plan free for K-12. This initiative seeds Figma’s usage among the next generation of designers and product builders, creating early familiarity and loyalty

• Strong Community efforts: Community has also been a driver of growth. Over 250,000 community resources and 10,000+ plugins/widgets help users customize and extend Figma. There are over 200 Friends of Figma chapters worldwide, from Lagos to Los Angeles which host over 650+ events globally.

Over time, Figma has complemented its core PLG motion with a stronger sales-driven approach. Figma started adding dedicated sales personnel around 2018, marking a pivotal shift toward enterprise adoption.Approximately 70% of its revenue now comes from Organization and Enterprise plans, demonstrating significant traction among larger enterprises. This blend of bottom-up PLG and targeted enterprise sales significantly enhances Figma’s market strategy and scalability.

The AI Wildcard

By all accounts, Figma is an extremely strong business: retentive, growing well, and profitable. But perhaps the only real “gotcha” for Figma is the question around AI.

AI represents both a major risk and a large opportunity, as well as the most significant uncertainty facing Figma. The rapidly evolving generative AI landscape is already starting to change the way the product development process works.

With the rise of tools such as v0, or vibecoding tools such as Lovable and Bolt, some people are leaving design to the AI for at least the lower-tier of personal use/internal use applications, and doing it as part of the integrated process.

The company openly acknowledges this AI risk and that AI could commoditize parts of the design process, potentially impacting pricing power and user engagement.

[AI has the potential to] alter.. how consumers and businesses interact with websites and apps and consume content in ways that may result in a reduction in the overall value of interface design, or by otherwise making aspects of our platform obsolete or decreasing the number of designers, developers, and other collaborators that utilize our platform

At the same time, AI represents a significant opportunity for Figma as well. Figma has proactively integrated AI capabilities across its product suite. Offerings like Figma Make leverage AI for faster prototyping, with management positioning Figma as the central collaborative platform even as AI automates routine tasks.

The company talks about the potential of AI to lower the floor and raise the ceiling in the design process, which is phrasing I quite liked:

With Figma Make, we believe even more people will be able to create functional, dynamic prototypes and software, as it lowers the floor to let absolutely anyone transform their idea into something that works. At the same time, it raises the ceiling of what’s possible by giving people access to capabilities that weren’t possible in their current tools, whether that’s allowing designers to tap into the power and expressiveness of code, or allowing developers to enjoy the freedom and speed of visual manipulation

The introduction of AI-driven tools may also increase short-term hosting and computational expenses, potentially impacting gross margins. Long-term, Figma's success in navigating the AI shift will hinge on how effectively it integrates these tools into its ecosystem without sacrificing pricing power or user retention.

Closing Thoughts

Figma’s IPO underscores the value of patience, product-led growth, and strategic evolution. Its financial metrics including rapid growth, impressive margins and capital efficiency are outstanding. The transition from a single-product design tool to a comprehensive, enterprise-focused suite has been exceptionally well-executed.

Nevertheless, the AI wildcard presents significant uncertainty, making Figma’s ability to adapt and innovate crucial to its future success. I'll be closely watching how the company continues to evolve its business model, manage enterprise adoption, and navigate the evolving AI landscape.

Overall, Figma stands out as a true generational company that began a lot of the PLG and cloud-based collaboration movement and one of the highest quality application layer SaaS companies going public we’ve seen recently.

Based on the quality of the business and the current comps on a Rule of X / NTM growth basis, I expect it to trade at an ARR multiple of ~19-22x which should put it at ~$18-20B range (they also have 1.5b in cash). I won’t be surprised if the IPO sees strong demand and so trades above that range from the get go either.

The company also deserves a lot of credit for the way they handled the post-Adobe failed acquisition state. The business was at $400M at the time the acquisition was announced. The company has remained focused and executed and more than doubled ARR over the next 2.5 years, and done so while being profitable and efficient and accelerating their multi-product vision and should be rewarded with a valuation around what they would have gotten with Adobe if not a bit more as they start trading.

Subscribe now

Hi friends,

Lately, there has been a lot of discussion among founders and investors about a renewed interest in “seed-strapping”. While this approach has always existed, we're seeing a dramatic increase in founders choosing this route, especially as AI makes it easier to scale with less capital.

Let's dive in and explore exactly what seed-strapping means, why it's become possible, and whether it's truly a sustainable model for building enduring companies.

What is seed-strapping

“Seed-strapping” is when a startup raises just one early funding round (usually a pre-seed or seed in the range of $500K-$4M) and then relies solely on revenue to fuel further growth. Unlike the traditional venture path, seed-strappers skip the constant fundraising but instead rely on the revenue and more specifically the profits on the revenue they generate to grow their business post achieving product market fit.

Think of it as bootstrapping with a head start. You get enough runway to prove product-market fit and achieve profitability, then you never look back.

But why the renewed interest in taking this route? The answer lies in how AI has changed the economics of scaling software companies.

How AI is fueling seed-strapping

Two powerful trends have made seed-strapping more achievable than ever:

A. Strong early product-market fit

AI tools have dramatically accelerated the path from idea to revenue. Teams of two to four engineers can now ship sophisticated products in weeks instead of months or years. In addition, the time-to-value of AI products can be quite quick and their ROI quite high, meaning that products built relatively quickly can be sold quite early, both through bottoms up motions as well as to businesses which all have mandates and deep interest in adopting AI.

Thanks to this rapid customer adoption coupled with faster build times, some startups are hitting millions in revenue with much less upfront capital.

B. Doing more with fewer people

Even post product-market fit, AI can help reduce the capital needed to scale a business significantly. AI can automate repetitive tasks across various departments such as customer support, sales and marketing. Coupled with the strong ROI on AI products in R&D functions such as engineering and design, startups can even continue to scale much faster with small teams.

’s Lean AI Leaderboard highlights dozens of companies that have grown to 10s of millions of revenue burning under a few million and with 20-30 people. These metrics were extremely rare if not almost unimaginable just a few years ago. While many of them may not seed-strap in the definitional sense, seeing what is possible with small teams and amounts of capital encourages other to potentially go that route, at least initially.

Who's actually seed-strapping

While the term has become popular relatively recently, there are a few examples of the phenomenon going back:

• Zapier raised just $1.3 million in 2012, reached profitability in 2014, and is now valued in the billions, thus far not raising any other primary capital.

• Calendly raised a modest $550k seed round in 2014 and operated profitably for years before eventually raising a very large $350 million “Series B” at a $3 billion valuation.

• While not quite seed-strapped, Veeva raised $7M in seed and Series A money in 2007 and 2008, before not raising again until their IPO in 2013

Recent AI-native startups like Aragon (10M runrate, <1M raised) and Jenni.ai (10M runrate, <1M raised) and Pump (15M runrate, <5M raised) are all examples which so far have scaled to double digit millions of ARR and only raised one round so far, although of course they may decide to raise another round in the future.

If you don’t yet receive Tanay's newsletter in your email inbox, please join the 10,000+ subscribers who do:

Subscribe now

Why seed-strapping is appealing and risky

The allure of seed-strapping to founders is quite clear:

• Founder control: Less dilution means founders retain more equity

• Operational discipline: Without deep pockets, companies stay lean and more disciplined

• Strategic optionality: When your growth is revenue-funded, you control if and when you raise additional capital opening up a broader range of outcomes in which everyone involved with the company does well as well as paths for the companies (remain profitable and distribute dividends, etc)

• Hiring: For certain people, seed-strapped startups offer less risk and potentially more stable work environments. The below comment on Hacker News also highlights how it can be appealing to prospective hires since it de-risks companies in terms of runway and certain macro factors.

At the same time, the risks are real, and it may only work in certain industries and markets.

• Losing out to better-funded players: This is perhaps the biggest risk of seed-strapping. It’s very possible that a better funded competitor raises more money and leverages that to hire more people, build a better product, be more aggressive on GTM and crush the seed-strapped competitor. Parker Conrad (CEO at Rippling) encapsulated this danger well:

  

• May put a ceiling on growth: Ultimately, if a company is operating in a large market and can deploy capital in a way that the return on that invested capital is high, it's in the interest of the company to invest additional capital. Being dogmatic about seed-strapping can sometimes restrict the ability of the company to grow as quickly as it could be or grow as large as it could have been otherwise.

• Hiring challenges: Without splashy valuations and large equity pools, seed-strappers may struggle to recruit talent in certain cases, particularly depending on the talent it needs.

• Limited liquidity: Early investors (and founders and employees for that matter) may find themselves stuck as it relates to liquidity.

The reality is, seed-strapping isn't foolproof. It's powerful but comes with meaningful trade-offs.

When does seed-strapping work

Seed-strapping works best when:

• Early product-market fit: You can hit meaningful revenue inside a year, so the seed cheque stretches until the business funds itself.

• Low-cost distribution: Product-led growth, usage-based pricing, or self-serve onboarding keep CAC small and scalable.

• High margins and fast payback: Cloud delivery, 70%+ gross margin, and sub-12-month CAC payback turn revenue into working capital.

• Fragmented or niche markets: Multiple winners can thrive, so depth of execution beats speed of fundraising.

It's a poor choice for:

• Winner-take-most markets: Marketplaces, social graphs, and broad horizontal platforms reward whoever scales fastest.

• Heavy upfront spend: Hardware, biotech, deep infra or regulated fintech need large R&D or compliance budgets long before revenue.

• Capital as a trust signal: Seed-strapping can be difficult when buyers or talent look for companies with strong balance sheets, visible reserves for support, and the ability to fund continuous innovation.

• Outsized ambitions in tight races: When the market is hot and every incremental dollar can be reinvested at a high return on invested capital, a larger war chest usually beats thrift.

The "skip-the-A" variant

There's another intriguing variant of seed-strapping emerging: the "skip-the-A" strategy.

Here, startups raise their pre-seed/seed round but then stay lean enough to bypass the traditional Series A altogether, jumping directly into larger rounds (like a Series B or C) at higher valuations typically in the double digit millions of ARR.

These rounds may still be labeled as a Series A when they happen, but look more like a growth round (Series B/C) in their size and valuation.

Garry Tan from Y Combinator points out how this is becoming more common:

Calendly mentioned above famously executed this playbook, and it's becoming increasingly appealing to AI startups with explosive early revenue where given a combination of factors such as strong ROI, explosive prosumer adoption, they can scale to 8 figures in ARR very quickly.

8. Closing thoughts

I believe seed-strapping is a real option worth considering for some AI-native startups. But like all strategies, it's neither perfect nor universally applicable.

Ultimately, whether seed-strapping thrives or fades will depend on how effectively founders navigate these trade-offs. The winners will be those who correctly identify which path matches their business model. I don’t think it threatens the typical venture model soon since thus far the very best companies have continued to raise capital, but may change the dynamics of what future rounds look like and exit ownerships in many cases, as it represents a sign of the idea that companies can accomplish more and more with less and less capital.

Subscribe now

Hi friends,

Stablecoins have quietly grown from crypto curiosities into genuine payments rails. Circle’s recent S-1 and subsequent public offering offers a unusually detailed view into how a public-company stablecoin operator works. Let’s dive into some of the key takeaways from it.

1. Stablecoins are already a systemic payments rail

The two clear leaders in stablecoins are Tether’s USDT and Circle’s USDC. USDT still tops the league tables at roughly $150 billion in circulation, lifted by offshore trading desks and emerging-market demand. USDC is smaller at ~$61 billion outstanding, yet it settles value in more regulated venues and has already moved >$25 trillion on-chain since launch, including ~$6 trillion in Q1 2025 alone.

Circle’s token now moves value on a par with mid-tier national clearing systems and more than companies like Paypal.

2. Circle’s revenue engine resembles a floating-rate bond fund

Circle has a pretty simple business model: it parks the dollars that back each USDC in short-term U.S. Treasuries and keeps the interest those T-bills generate of which today there is over $60B. Rate hikes have therefore turbocharged revenue, although USDC in circulation has been a bit lumpy.

More than 97 percent of revenue comes from this interest, which they call as reserve income. Other revenue streams include include the small slice of transaction and infrastructure fees it charges businesses for using its payments, custody, and treasury APIs which are quite immaterial today.

The company’s revenue is thus very dependent on the reserve rate which thye are able to generate which is essentially a slight discount to the prevailing Secured Overnight Financing Rate (SOFR), which so far has gone in their favor the past few years, but if interest rates were to meaningfully decrease could hurt them substantially.

3. Coinbase is both a rainmaker and toll-keeper

Coinbase is Circle’s rain-maker and toll-keeper in one. Back in 2018 the two companies launched the Centre Consortium, splitting governance 50–50 and sharing interest income in proportion to how much USDC each one issued or custodied. That “you earn what you distribute” formula lit the fuse on USDC adoption without forcing Coinbase to guarantee any minimum volume.

Fast-forward to August 2023. Centre dissolves, Circle takes full control of policy, and Coinbase walks away with a minority equity stake plus a perpetual licence to use Circle’s stable-coin trademarks with a new three-year Collaboration Agreement now setting the money flow:

Circle skims a small issuer-retention fee to cover compliance and treasury. The leftover reserve income is shared in two layers:

1. Party-product slice: Each firm gets a cut that matches the share of USDC sitting in its own wallets and custody accounts.

2. Ecosystem slice: Whatever remains is split 50/50 as long as Coinbase keeps USDC easy to buy, supports it across key products, and shows up on policy.

The 2024 numbers show how big that lever has become. Circle booked about $1.7 billion of revenue yet paid a little over $1 billion back out as distribution costs, nearly $908 million of which went to Coinbase.

One-fifth of the entire USDC float now lives inside Coinbase’s ecosystem, and the exchange pushes velocity further through Base L2, gas-free smart wallets, and fee rebates. That distribution muscle is priceless for growth, but it also means upward of 50%+ of Circle’s revenue will go to Coinbase likely into perpetuity.

4. Institutional partnerships deepen credibility

In addition to Coinbase, two other partnerships can prove critical to Circle’s continued growth.

Binance for non-US growth and circulation

A November 2024 agreement turns the world’s largest exchange into a growth engine for USDC. Circle paid Binance a one-time fee of $60.25 million and now sends monthly incentives that rise with the exchange’s USDC balances. Binance has pledged to keep at least $1.5 billion of the stablecoin in its own treasury, with a working goal of $3 billion, and to market USDC to the platform’s 240 million users.

BlackRock for credibility and asset safety

An initial 2022 partnership made the firm exclusive manager of USDC’s Treasury portfolio. A new four-year agreement signed in March 2025 tightened the link. Circle now commits to hold at least 90 percent of its U.S. fiat reserves (excluding bank deposits) in BlackRock vehicles, including the SEC-registered Circle Reserve Fund. In exchange BlackRock agrees to prioritise USDC in its own capital-markets work and to stay out of launching a competing dollar stablecoin. BlackRock also took a 10% stake in Circle’s IPO.

5. The IPO Pop shows Demand for Stablecoin Pure-plays

Circle priced its IPO at $31 a share and raised about $1.05 billion, then finished its first session at $83.23. The stock has since crossed triple digits, representing a market cap upwards of >25B. How its traded has sent the clearest possible signal that public-market investors wanted direct exposure to the stablecoin theme.

Annualising first-quarter numbers puts 2025 revenue around $2.3 billion, adjusted EBITDA near $490 million, and net income roughly $260 million. On those run-rate figures the market is paying about eight times sales, forty-seven times EBITDA, and ninety times earnings.

For comparison Visa trades near fifteen times EBITDA but carries a mature five-percent top-line growth profile; money-market funds carry virtually no multiple at all.

It’s a sporty multiple for a company still navigating real risks: nearly all of Circle’s revenue comes from interest on reserves, which could compress if rates fall. More than half of that revenue is paid out to Coinbase in distribution costs. And despite years of effort, the company hasn’t meaningfully diversified beyond interest income.

But the market seems to be betting that Circle can thread the needle. Investors are underwriting continued double-digit growth in USDC circulation, meaningful traction in fee-based products like CCTP and merchant APIs, and a regulatory glide path that could entrench USDC as the compliant dollar of choice.

Subscribe now

Hi friends,

Last time, I wrote about how AI agents are accelerating adoption cycles for developer tools and choosing tech stacks. But agents will play a role in the distribution of much more than just software products. They’ll fundamentally change how all commerce is done over time, and today I want to explore that idea: agentic commerce.

I’ll cover:

• What agentic commerce is

• The current state of agentic commerce

• Opportunities in agentic commerce

What is Agentic Commerce?

eCommerce already accounts for about 20% of all retail sales, representing nearly $6 trillion in transaction volume. Today, consumers research products across the web, consult friends and family, engage with social platforms, and browse retailer websites to discover and narrow down their options. They then complete purchases through marketplaces or online stores.

Over time, each part of this funnel has evolved. For example, 50% of consumers now start product searches on Amazon rather than Google. For Gen Z, discovery has shifted even further towards platforms like TikTok and Instagram.

Agentic commerce is the next step in that evolution. It extends this shift across the entire funnel, with AI agents rather than humans handling discovery, comparison, negotiation, and purchase.

Even today, consumers are already using tools like ChatGPT to research and narrow down product choices. And this is even before these tools had fully integrated memory, personalization, or been tailored for shopping experiences.

As these experiences mature, imagine asking your personal AI, “Find me running shoes under $120,” and instantly receiving tailored recommendations based on your past preferences and sizing. The agent identifies the best price, selects the right retailer, and completes the purchase with no further input required.

The graphic below outlines how this buying journey could evolve, as consumers increasingly rely on personal AI agents to discover, evaluate, and purchase products potentially over time even proactively, before the consumer even asks.

Current State of Agentic Commerce

As discussed above, consumers are already turning to AI products such as ChatGPT for product discovery (with 42% of users in the Bain survey indicating that they have used it for product discovery). But beyond that, we're seeing signs that the key platforms are going deeper into agentic commerce and taking this use case more seriously, even though agentic commerce is very much still in its infancy.

A. OpenAI Shopping

OpenAI announced a shopping initiative where users will see enriched product information for product related queries and see options to purchase the product. For now, users will still be redirected to the merchant’s website to complete the transaction.

B. Google Agentic Checkout

Google at I/O announced that the new AI mode of search will integrate their Shopping Graph (which has a catalog of 50B products) and allow for browsing and discovery of products. It will then also be able to use agentic checkout to buy the products on behalf of the consumer.

C. Perplexity Shopping

Perplexity improved their product discovery a few months ago with enriched listings and added support for a “Buy with Pro” feature for pro users for some SKUs where the checkout happens natively within perplexity.

Overall, these tools are just scratching the surface around product discovery and checkout given that most SKUs don’t support it, and the native experiences for discovery support only a subset of products and are still very much in their early stages.

If you don’t yet receive Tanay's newsletter in your email inbox, please join the 10,000+ subscribers who do:

Subscribe now

Opportunities in Agentic Commerce

At a high level, I think of the agentic commerce stack as below, comprised of six key types of companies: broad agentic platforms and verticalised commerce platforms that provide consumer shopping experiences; and catalog infrastructure, transactions and payments, merchant tools, and post-purchase experiences that enable these experiences for merchants and consumers.

1. Broad Agentic Platforms

Broad agentic platforms will serve as the general-purpose entry point for consumers interacting with AI-driven shopping. These platforms aggregate consumer intent, handling diverse product requests across many categories. Their primary role is to translate user intent accurately into actionable shopping experiences. Over time, I expect these platforms push deeper into the commerce experience, potentially relying on 3rd parties or by creating standards that merchants will then adhere to given their scale. For example, Perplexity and OpenAI are experimenting with letting merchants directly submit detailed catalog data.

Examples: ChatGPT, Claude, Perplexity, Meta AI

2. Vertical Commerce Platforms

Vertical platforms that specialize in specific industries or product categories, offering deeper insights and more tailored shopping experiences for that vertical are likely to also emerge and play a role. These platforms will be commerce centric and leverage specialized data to provide better recommendations, detailed insights, and industry-specific expertise, significantly enhancing customer satisfaction and loyalty.

Examples: Amazon Rufus (1P Amazon products only), Google Shopping, Daydream, Phia (focus on comparison agents across new and secondhand), Onton (furniture/home decor oriented)

3. Product Catalog Infrastructure

Product catalog data, particularly in a form that can be read by agents is a critical part of the stack and ensures ensures accurate, structured, and real-time product information is available to agents. Better data when coupled with information about consumer preferences from the agent will allow for better discovery and recommendations as well as surfacing products actually in stock so that transactions can take place.

It remains to be seen whether the way this plays out is that providers that aggregate across merchants and sell to the agents emerge, or if the agent platforms create a standard and all merchants over time try to abide by that standard, offering an agent-ready version of their catalog/storefront.

Examples: Companies like Zinc, Rye, Velou, Shopify Catalog all offer agent-ready catalogs/APIs that can be consumed by agents.

4. Transactions and Payments

Historically, bots have been considered bad in payment and checkout flows since they indicate fraud. Now, with agentic commerce, agents that are given permission on behalf of users to purchase items for them means that not all bots are bad, and the checkout and payment flows need to adapt accordingly. In addition, agents need to be able to reliably navigate and checkout on stores, which today were not built for agents. So we’ll see some transactions and payments infrastructure that streamlines the checkout and purchasing process, enabling secure and efficient transactions initiated by AI agents.

Stripe has already demo’ed their Order Intents API as an example of this, and Visa has an entire initiative around intelligent commerce that touches on agentic payments.

Examples: Companies like Firmly, Stripe, Nekuda, Skyfire are all addressing various aspects of the checkout and payments flow.

5. Merchant Visibility & Optimisation

As AI agents increasingly mediate between brands and consumers, merchants require tools to optimize their visibility within AI-generated recommendations. This includes analytics and insights into how products are presented and ranked, helping brands effectively tailor their offerings to AI-driven discovery, as well as potential shifts in what the concept of a “storefront” optimised for AI agents mean. This goes beyond commerce and will be needed by all kinds of companies, but in commerce there may be additional needs around catalog and payments as touched on above.

Examples: Companies such as Profound, Bluefish, Evertune and Gumshoe are all building products to help merchants (and other kinds of companies) monitor and improve their visibility in AI engines.

6. Post-purchase Experience

If as you play it out fully ChatGPT and other agentic products are where people are making purchases, that's also where they'll go to facilitate returns or other kinds of customer support. For this reason, we'll probably see the need for post-purchase infrastructure to be made available to these agents to manage returns, customer service, track shipments and other tasks (given that that's where the consumer will be expecting to take these actions).

Examples: Companies such as Loop and Aftership support API-driven returns and shipping tracking which can be made available to AI agentic platforms. In addition, companies such as Gorgias and Sierra assist with support for merchants, and one can see them making their support available directly through the agentic platforms if that’s where consumer attention is (rather than through the merchant websites).

Closing Thoughts

Agentic commerce is rapidly approaching. Initial steps by major companies such as Google, Stripe, Visa, and OpenAI point towards substantial infrastructure growth. Yet, many critical components such as robust product data APIs, streamlined checkout experiences, vertical experiences, and visibility tools remain open opportunities for innovative founders.

We’ve optimized storefronts for human attention and clicks. But as AI agents take over the shopping journey, the question becomes: what does a storefront designed for machines look like?

If you’re building in and around the area or thinking about the area I'd love to hear from you. Please feel free to reach out to me on Tanay @ wing.vc.

Subscribe now

Hi friends,

As part of Databricks’ acquisition of Neon, they included a jaw-dropping chart: 80% of new databases spun up last month on Neon were created by AI agents, not people.

The same pattern is spreading as more and more non-developers use “vibe-coding” platforms such as v0, Replit, Bolt, Lovable all of which are on a tear. Non-developers put a prompt and an AI agent expands the project requirements and generates the code and the integrations baked into that template become the long-term defaults for core systems such as auth, payments, databases, etc.

Even with tools like Cursor, which targets engineers, as more of the code is written by coding agents, some developer tool and infrastructure decisions are being pushed to the agent as part of an implementation detail (based on the model being used under the hood). Once the agent picks a tool, humans may not bother revisiting the decision unless something breaks.

The Stack of Choice for AI Agentic Tools

The graphic below shows which sets of tools1 and products are being used as defaults by the various AI app builders and vibecoding platforms. As you can see, users end up using many of the same set of tools by default: Neon or Supabase for databases, Vercel or Netlify for hosting, Stripe for payments, Resend or SendGrid for email. The picture explains why some vendors are suddenly everywhere: the app builders with end adoption are creating more and more pieces of software, and they are choosing these tools (thanks to integrations and templates) as the defaults.

As you can see, some also preference their native tools for various components of the stack, most visible in the case of v0 by Vercel and Replit agent by Replit.

Four Concrete Steps for Dev‑Tool Builders

1. Optimize for LLM Visibility

Large language models ultimately drive these tools and agents and monitoring and being visible to them is critical. They suggest code snippets based on what they have seen in docs, GitHub repos, public Q&A, and official API specs. If your product is invisible to the model, the agent never considers you.

What to do

• Publish a clean OpenAPI or GraphQL schema. Models parse structure better than prose.

• Seed high-quality, copy-pasteable examples across GitHub, Stack Overflow, and your own docs.

• Track how you show up in LLMs using an LLM visibility monitoring tool

Example: Upstash’s Redis client appears in thousands of Next.js repos. That density makes “add caching” tend to auto-complete to Upstash inside v0, despite no official integrations.

2. Ship Zero-Click Provisioning and Guardrails

Agents will not navigate dashboards or paste keys manually. They need a single HTTP request that creates a resource, returns credentials, and shuts down cleanly when idle.

What to do

• Let a single API request spin up a new project and return its URL, username, and password in under a second.

• Issue narrowly scoped service tokens so the agent never has to handle the master credential.

• Make it just as easy to clean up: one call should delete the project when the demo ends.

• Default to scale-to-zero so unused sandboxes cost nothing.

Example: Neon delivers in this regard: a new Postgres branch is ready in ~800 ms, costs pennies while idle, and can be forked for every test run, which allowed it to become the default database for Replit and v0.

3. Become the first-click integration inside agentic builders

Defaults and integrations with the prominent AI app builders is critical to winning and capturing these new applications and workloads. One first-party template can outperform 100s of blog posts.

What to do

• Build official starters for each editor that import your SDK, inject env vars, run a health check, and commit a sample file.

• Include documentation for non-developers on how they can integrate your product with their app builder

• Keep the template owned by your team so you can push fixes on release day.

Examples: Stripe wrote native components for v0. When a user asks “add payments”, the agent pastes a ready-to-run checkout form instead of a link to Stripe docs. Payments on other processors require manual wiring, so users rarely switch.

Similarly, Posthog has instructions for non-developers on how to integrate Posthog with these AI agents, despite not having official integrations yet or being the official default yet.

Closing Thoughts

The procurement funnel is shrinking to a single function call. When an AI agent can spin up a database, cache, or payment gateway in under a second, it locks in that choice long before a human engineer even opens the repo. That shift raises the bar for every dev-tool company: your marketing site is now your OpenAPI spec, your growth loop is public code examples, and a critical distribution channel is whether the agents and app builders use you as a default or recommended integration.

Make your product trivial for a model to reference, instant for an agent to provision, and obvious for a no‑code user to trust. Do that and your logo will show up in templates, spread with every prompt to app “generate” click, and become the default the next generation of builders learns first. Build for the AI agent first and the developer will follow.