partner @ Accel. head cheerleader @ gamma, lovable, radar, ironclad, transcend, permitflow. formerly: frame (ADBE), spoke (OKTA), segment (TWLO), deepmap (NVDA)

Joined April 2009
49 Photos and videos
The path from raw inputs to AI end-user value is long and constantly mutating. Every layer has its own supply-demand curve with mid-cycle attempts at reinvention (e.g. grid power > on-site turbines, copper > photonics, HBM > next-gen memory stacking). Yes, the aggregate AI infra curve is steep. But alpha lives in the component curves: the companies, their individual dynamics, how they're financed, where the market still underprices forward demand… There's no grand unified theory. But two axioms apply to each sub-market: 1. Cost of capital is a thing. You create equity value when returns on invested capital exceed financing costs. Growth funded below that hurdle destroys value, no matter how big the TAM. That's the gravitational force that pulls us back down from lofty "revenue multiples." 2. Cyclicality is a thing. The second the market senses overshoot, multiples collapse — and you can be left painfully holding the bag. But combing through this tangle is the fun part.
1
2
122
"The gap between 'model can do it' and 'user gets value from it' is still large." - Anton Osika It's the intense focus on compressing that gap that underwrites @Lovable's amazing growth story. $500m in revenue! Bravo.
50M projects built. 720M monthly visits to apps made on Lovable. 80% of builders are non-technical. 35% are already generating revenue. The numbers prove a shift: this is the build economy. thebuildeconomy.lovable.app
4
646
Cloudflare x @voidzerodev! Agents loop alongside AI to produce more and better software -- arguably the most important capability to emerge in the last few years. Vite is the build tool beneath it all. Big congrats to Evan and the team In other news @caseyaylward @Accel is on a tear. I'm a very proud Indian uncle. blog.cloudflare.com/voidzero…
9
556
Two interesting and (converging?) paths in applied AI land. 1. Land with an AI-as-intelligence product that maps to use cases in deep knowledge work domains, where reasoning about primary sources can be challenging. Fast time-to-value, good ACVs. Rapid growth. 2. Decompose a business process and re-encode via AI. Grind through change management to automate workflow. Lots of multi-step processes and presuming there's low error tolerance, startups can differentiate on verifiable or traceable work. Not quite the growth rates of bucket 1, more "deliberate" growth -- but potentially long-term stickier. Former are moving in the direction of the latter (whether explicitly or not) -- JVs, acquiring customers or building pseudo-BPOs. Latter are building the operational IP to help customers "get to outcomes". Ultimate question is where equity value accrues. The pie chart near term may heavily favor type 1 companies. But may rotate over to type 2 in the medium-to-long term. Fun reality is the pie itself appears so large that both can win. But the slices themselves are changing texture real-time.
1
1
13
1,843
Decompose most AI infra startups and the moat resolves to supply-chain mastery — not just access to components, but combining or operating them. From a cold start that's a difficult layer to accurately underwrite. Pull the thread and diligence runs into fairly opaque corners of the market.
1
7
1,305
Watching the memory trade go parabolic -- curious how the market resolves some key questions. How does demand adjust to current lead times? What's the "phantom" backlog? To what extent is backlog serviced by cheaper substitutes if delivery slips? Pushing delivery out a year or two isn't the issue -- it's the risk demand evaporates before it can be fulfilled. Markets seem to be pricing near permanent, inelastic demand for high-bandwidth options. But memory has always been a cyclical semi.
345
We see massive spend attached to modernizing/maintaining/migrating existing software -- an ideal problem for AI to reason about. Real value if you can help enterprises manage what they already have versus selling them on something new. You'll see this more in the @Accel portfolio -- @nova_ai , @Code_Metal_AI , Agave and others.... startups that run toward domain-specific complexity to help customers traverse their sunk cost fallacy from prior builds.
We @Accel are excited to have led the seed round in @getnova_ai, and deepen our partnership with @EmmaQian_, @samuely4ng, and @dr_zeier as they announce their $31.5M Series A (and $40M in total funding), led by our friends at @chemistry (@kshenster). Nova is building the leading AI development platform that enables enterprises to generate and transform complex software using natural language - starting with SAP. Instead of writing code line by line, teams can describe problems, workflows, and business logic directly, and Nova instantly translates it into working implementations. This also means everybody on the team can now engage with code, from engineers to functional analysts, resulting in faster development cycles, lower costs, and a fundamentally more scalable approach to enterprise software adoption. We believe this is the beginning of a much larger shift. As AI moves deeper into the enterprise, the opportunity is not just to accelerate coding, but to redefine how business systems are built, customized, and maintained. Hear more about the vision directly from @EmmaQian_ and @samuely4ng
1
7
738
Lots of focus on AI building new software. Increasingly white-hot market is how AI operates atop or between existing software.
1
125
vas natarajan retweeted
We @Accel are excited to be leading @RadixArk’s $100M seed round alongside our friends at @sparkcapital. It’s been a privilege to partner with @ying11231 and @BanghuaZ since inception and to support the incredible @lmsysorg community. Intelligence is no longer the bottleneck! Developers are now constrained by their ability to control, adapt, and reliably serve a growing diversity of AI models across hardware, environments, and use cases at scale. This shift creates an opportunity to build new foundational infrastructure for training and inference. RadixArk’s mission is to build that open infrastructure, and we’re excited to be their partners on the journey.
Today, we are thrilled to officially launch RadixArk with $100M in Seed funding at a $400M valuation. The round was led by @Accel and co-led by @sparkcapital. RadixArk exists to make frontier AI infrastructure open and accessible to everyone. Today, the systems behind the most capable AI models are concentrated in a small number of companies. As a result, most AI teams are forced to rebuild training and inference stacks from scratch, duplicating the same infrastructure work instead of focusing on new models, products, and ideas. RadixArk was founded to change that. We are building an AI platform that makes it easier for teams to train and serve the best models at scale. RadixArk comes from the open-source community. We started with SGLang, where many of us are core developers and maintainers, and expanded our work to Miles for large-scale RL and post-training. We will continue contributing to both projects and working with the community to make them the strongest open-source infrastructure foundations for frontier AI. We would like to thank our long-term partners, contributors, and the broader SGLang community for believing in this mission. We're also grateful to @Accel and @sparkcapital, NVentures (Venture capital arm of @nvidia), Salience Capital, A&E Investment, @HOFCapital, @walden_catalyst, @AMD, LDVP, WTT Fubon Family, @MediaTek, Vocal Ventures, @Sky9Capital and our angel investors @ibab, @LipBuTan1, Hock Tan, @johnschulman2, @soumithchintala, @lilianweng, @oliveur, @Thom_Wolf, @LiamFedus, @robertnishihara, @ericzelikman, @OfficialLoganK, and @multiply_matrix among others. Thanks for the exclusive interview with @MeghanBobrowsky at @WSJ about our vision.
3
10
39
5,853
vas natarajan retweeted
Lovable is in the Forbes AI 50 this year. We’re honored by the recognition. It’s motivation to keep building. Now, back to work.
16
17
244
15,788
vas natarajan retweeted
Lovable is now the easiest place on the internet to create something and then make money from it. It can cover any use case. For example you can create an ecommerce store selling physical products, or sell software with a subscription. You can do this with an existing company, or fully seamlessly using a merchant of record setup, where all complexities such as global tax is handled for you.
Apr 13
Introducing Lovable Payments. Describe what you want to sell. Test it securely. One conversation to go live.
14
10
115
18,228
vas natarajan retweeted
Apr 7
Hello, Moon. It’s great to be back. Here’s a taste of what the Artemis II astronauts photographed during their flight around the Moon. Check out more photos from the mission: nasa.gov/artemis-ii-multimed…
9,921
173,097
807,236
29,918,757
Huge congrats to the Astral team! What an incredible group of builders and humans bringing deep thought and care to the Python tooling world. @charliermarsh And my partner @caseyaylward who saw it before anyone and led our Seed and Series A. “Python is the language of AI, and Astral can become a major Python company”.
Mar 19
@OpenAI announced today that it’s acquiring @astral_sh. This is a testament to @charliermarsh and the entire Astral team’s craft, dedication, and skill for building tools that developers love. Thank you to Charlie and to everyone at Astral for welcoming us as partners on this journey. We’re cheering you on in this next chapter. Read more from @caseyaylward about what Astral has accomplished and where they’re headed next. ⬇️ accel.com/noteworthies/astra…
16
2,071
Team @GammaApp takes every step behind how a company creatively shares and compresses it ever-so-perfectly into their product. When you stack together releases like this, you see a new default creative system being built before our eyes. Think about where this all goes. And run, don't walk to our careers page, 🙏!
There's a hidden tax on every knowledge worker in the world, and nobody talks about it: The design tax. You're a strategist, a sales lead, a marketer. You were hired for what you know. But every meeting, every pitch, every proposal expects you to show up with something that looks like a designer made it. I lived this. Before Gamma, I spent time in consulting and investment banking. I spent more hours formatting slides than the analysis that went into them. When my cofounders and I started Gamma, we asked: what if you never had to be a designer in the first place? Five years and nearly 100 million users later, we've refunded billions of hours of the design tax. Today, we're eliminating it for good with our biggest launch ever. Gamma Imagine — a powerful, AI-native visual creation tool directly in Gamma. Posters, logos, infographics, visuals from a single prompt. On brand, every time. AI-Native Templates. Templates were supposed to save you from design work. Instead you spent the time filling them in. So we completely rebuilt the template experience. Modify a whole deck with a single prompt, with your brand and style intact every time. Gamma Connectors. You're already thinking in ChatGPT and Claude. Now Gamma sits inside the most popular work apps in the world. No more context-switching. You were hired for your ideas, not to resize text boxes. Let Gamma pay the design tax.
1
5
621
When every person at a company has an insanely high standard for themself, the standards for the company take care of themselves. Incredible people x right mission @Lovable
Lovable says it added $100M in revenue last month alone, with just 146 employees techcrunch.com/2026/03/11/lo…
457
vas natarajan retweeted
Two years ago, we made the biggest bet in Gamma's history. AI models were getting dramatically better, and we decided to rebuild our entire product around them. People thought we were crazy, we were a small team, pre-revenue, and we were essentially starting over. That bet took us from a few thousand users to 50 million. From zero revenue to $100M ARR. It transformed how people create and share ideas. Now we're seeing the same kind of shift happening again. The world is going agentic and just like we went AI-first two years ago, we're now building Gamma for an agent-first world. Today we're launching the Gamma connector for Claude, in partnership with @AnthropicAI. When Claude finishes a task for you (a research summary, a project plan, meeting notes) it can now hand you a Gamma presentation built from that work. Your thinking becomes something visual, polished, and ready to share. This is our first step. We believe that as agents become the way people get work done, every agent should be able to express ideas visually. We want Gamma to be the presentation layer for the agentic world. We're just getting started.
74
51
840
113,326