We help the daring build legendary companies from idea to IPO and beyond.
Menlo Park, CA
Joined March 2009
- Tweets8.9K
- Following1.6K
- Followers821K
- Likes10.9K
Sequoia Capital retweeted
@pydantic just launched agent sandboxes. But it's not just another garden-variety sandbox! Instead of firing a full VM or container to run agent-written code, Pydantic Agent Sandboxes uses a minimal interpreter with milliseconds start time, and the ability to snapshot, pause and resume any time.
Many companies have started building their own sandboxes based on Pydantic Monty. No need to DIY anymore!
Go straight to the source!
🚀 Today we're launching Agent Sandboxes in @pydantic Logfire: a profoundly different kind of sandbox.
"The world doesn't need another sandbox" they say. Maybe. Or maybe the world doesn't need another firecracker/gVisor/VM infra-fest.
Our Agent Sandboxes are different. They're just an isolated interpreter with cold start times measured in milliseconds, where the code can only call the tools you give it and use the directories you mount.
A massive bonus is you can snapshot and resume a REPL weeks later, with the snapshots costing virtually nothing.
Programmatic tool calling, solved.
Start free, $1000 of free credit to try it in the first month 👇
pydantic.io/launch-agent-san…
Competition heated up at our second regional showdown for the No Limit Bot Building Tournament in Boston Monday night.
- 270 teams
- 330+ students
- 30 universities
It was an arms race. In round 1 the bots bid an average of 28 chips for the flop. By the final round it was 66.
5 winners are headed to the Finals in San Francisco this Sunday, 11/15. Want to come watch? Email us: pokerbot@sequoiacap.com
Thanks to our partners @OpenAI @citsecurities.
Sequoia Capital retweeted
nice-looking trophy and great competition!
couldn't get cfr working well enough in time, but tons of stats, a few good heuristics, and fast testing got the job done
see you in sf ♠♥♦♣
Announcing No Limit: a bot building tournament hosted by @Sequoia, @citsecurities and @OpenAI.
Teams of 1 or 2 get 60 minutes to build a pokerbot from scratch.
Agents play thousands of games across six scored rounds.
Open to university students and recent grads (‘25 or ‘26).
Regionals in Boston, Pittsburgh and the Bay Area in early November. Top five teams from each fly to the Finals. $100K grand prize.
RT @Konstantine: .@rox_ai is the fastest growing company in GTM.
It puts technology at the heart of revenue generation. Today, their custo…
This quoted post is unavailable.
Sequoia Capital retweeted
Proud to announce our partnership with @edelbyte_io .
AI agents can run 50 queries where a person runs one. The data stack shouldn't make them ration their curiosity.
Edelbyte, from the team that built Kusto at Microsoft, keeps search and analytics fast and affordable at any scale, in any cloud.
sequoiacap.com/article/partn…
Want to come to the No Limit Finals in San Francisco?
- Top 20 student teams across 800+ participants build pokerbots to compete for a $100K grand prize.
- @OpenAI’s @polynoamial joins us for a talk with @Alfred_Lin.
- Founders and top players from @Etched, @Factory, @appliedcompute, @DoorDash, @fal, @Profound, @getserval, @sailresearchco, @conwayinc and more compete to determine the best poker player in AI. Proceeds go to @khanacademy. There are still a few spots left for partner companies to compete.
- We also have a few guest list spots available for spectators.
Interested? Email pokerbot@sequoiacap.com
In partnership with @citsecurities and @OpenAI
🤖 Made with AI
Sequoia Capital retweeted
We’ve raised $73.5M, bringing total funding to more than $100M.
Huge thanks to existing investors @sequoia , @Lux_Capital and @starship_vc and new investors, including @BuckleyVentures, @convectivecap, @UncommonGP and @SalesforceVC.
They’re amazing partners and we’re fortunate to be on this journey with them!
Our first demonstration mission, planned for the second quarter of 2027, will test how our reflectors unfold in orbit.
We continue to build thoughtfully and will share research on how our technology interacts with the world around it soon.
Onward!
reflectorbital.com/blog-post…
Sequoia Capital retweeted
On a Friday night in the Bay Area: Some people hang, some grind and some build pokerbots while playing chess at @sequoia
@abhishekm1636 accepted the challenge of how many IMO medalists can fit in a room
Sequoia Capital retweeted
I'll be at the table for @fal at the No Limit Cup this Sunday, 11/15 in SF — the charity poker game at the Finals of No Limit: A Bot Building Tournament, with all proceeds going to @khanacademy. Thanks to @sequoia, @citsecurities and @OpenAI for having us.
Who’s the best poker player in AI?
We’re about to find out. The No Limit Charity Cup is Sunday, 11/15 in SF. $50K per seat.
Founders and top players from @Etched, @Factory, @appliedcompute, @DoorDash, @fal, @Profound, @getserval, @sailresearchco, @conwayinc and more go head to head, with proceeds going to @khanacademy.
Think you’ve got what it takes? We still have a few seats for partner companies to send a delegate to compete. Want to come watch? DM me. Bragging rights on the line.
Thanks to our partners: @citsecurities @OpenAI.
Sequoia Capital retweeted
Fearing for those that have to compete with @dyu1729
Who’s the best poker player in AI?
We’re about to find out. The No Limit Charity Cup is Sunday, 11/15 in SF. $50K per seat.
Founders and top players from @Etched, @Factory, @appliedcompute, @DoorDash, @fal, @Profound, @getserval, @sailresearchco, @conwayinc and more go head to head, with proceeds going to @khanacademy.
Think you’ve got what it takes? We still have a few seats for partner companies to send a delegate to compete. Want to come watch? DM me. Bragging rights on the line.
Thanks to our partners: @citsecurities @OpenAI.
Sequoia Capital retweeted
Who’s the best poker player in AI?
We’re about to find out. The No Limit Charity Cup is Sunday, 11/15 in SF. $50K per seat.
Founders and top players from @Etched, @Factory, @appliedcompute, @DoorDash, @fal, @Profound, @getserval, @sailresearchco, @conwayinc and more go head to head, with proceeds going to @khanacademy.
Think you’ve got what it takes? We still have a few seats for partner companies to send a delegate to compete. Want to come watch? DM me. Bragging rights on the line.
Thanks to our partners: @citsecurities @OpenAI.
Sequoia Capital retweeted
Defense, space, energy, and robotics are all pulling hardware toward software-speed iteration. The teams winning in these markets ship in hours, days, and weeks, not the months and years hardware has traditionally taken.
Testing is where that pace breaks down. Test data lives in too many places and gets stitched together by hand. @nominal_io is the modern stack for this. They combine test automation, telemetry, and analysis in one place, from the first prototype through production and into operation.
Customers’ behavior is the best evidence. Nominal was recently selected for a $71M STRATFI, and won a delivery order award through PAE’s Rapid Capabilities Cell to speed up integration and testing of Castelion’s Blackbeard on the F/A-18. These are programs where test speed decides how fast a capability reaches the field.
Congrats to @CameronLMcCord, @jrshoch, @brycestrauss and the whole team on the Series C. We’re excited to keep partnering with them.
Nominal just raised $240M at a $3B valuation.
Engineers today are building some of the most ambitious machines humanity has ever seen. They should have AI that’s purpose-built for them.
We started Nominal with a simple observation, that hardware teams needed better ways to work with test data and make decisions from it. Over the past four years, we’ve built software that understands the complex realities of hardware development.
Now we’re building the Hardware Intelligence Platform that gives engineers increased visibility into complex physical systems, helps them make better decisions, and ultimately gets their products into the real world faster. It brings together the systems, data, intelligence, and applications hardware engineering requires to meet the AI moment.
Our Series C was led by @Lightspeedvp and co-led by @Insightpartners, with participation from Lux Capital, Sequoia, General Catalyst, Founders Fund, Avenir, Lockheed Martin Ventures, Salesforce Ventures, M13, DE Shaw Ventures, Deployment Capital, Avra, Red Glass Ventures, Carpenter Capital, Audacious, PSP, Overmatch Ventures, XYZ, Haystack, and Brave Capital.
Thank you to our customers, partners, team, and every hardware engineer building on Nominal.
Read more here → nominal.io/blog/series-c
Sequoia Capital retweeted
AI has supercharged software engineers. Nominal is doing the same for hardware engineers.
It's been a privilege watching this team make executing something incredibly complex look easy: multi-product, multi-industry, multi-segment, multi-use-case and multi-geo.
True smooth operators. Congrats to @CameronLMcCord, @jrshoch, @brycestrauss, and the entire @Nominal_io team. We @sequoia are proud to be your partners.
All systems Nominal!
Today we’re announcing @Nominal_io's $240M Series C at a $3B valuation led by @lightspeedvp and co-lead by @insightpartners. We’re also introducing the Hardware Intelligence Platform.
Over the last year, we’ve worked alongside our customers to answer one important question: What does it take for AI agents to earn their place in hardware engineering?
We take that question and the implications seriously. Hardware operates in the physical world, where the stakes are high, the constraints are real, and getting things right really matters. Moreover, the delta between the AI frontier and what many of our customers have available to them on a day-to-day basis in the industrial sector could not be wider than it is today. To date, agents, models, and purpose-built infrastructure haven’t been built with their use-cases in mind. In a world where context and data is spread across machines, files, drawings, spec sheets, test ranges, manufacturing lines, and most importantly, the brains of the very human operators, engineers, and technicians that power the physical world.
That changes now, with the first purpose built intelligence platform for hardware engineering. We're excited to show you what we've been building.
First, to our customers: Thank you for trusting us. This round lets us go further, faster, for each of you. For some, that means expanding beyond testing into design, production, and service operations in the field. For others, it means bringing agents into your daily work to truly operationalize AI in your organizations for the first time. Wherever you are in that journey, we're ready to meet you where you’re at, and build shoulder-to-shoulder as you take on the constraints you face every day.
Next, to our team: I'm very proud of what we've already accomplished, and now we have even more work to do. I’ve been inspired watching us move faster than ever, even while growing more quickly than at any point in our history. We started Nominal to redefine what’s possible in the physical world. That ambition hasn't changed, but the opportunity in front of us has grown enormously. The responsibility that comes with it isn't lost on us, and there's never been a better time to be a Nomineer.
And finally, to our investors: Thank you for continuing to believe in what we're building and backing us as our ambitions have grown. We began four years ago with redefining data analysis for hardware testing. That grew into full-stack data and control infrastructure for test and operations, and it’s now the foundation for an intelligence platform for the entire hardware lifecycle, from first simulation to fleets in the field.
The future of hardware will be built on Nominal.
All Systems Nominal 🚀
Lux Capital, Sequoia, General Catalyst, Founders Fund, Avenir, Lockheed Martin Ventures, Salesforce Ventures, M13, D.E. Shaw Ventures, Deployment Capital, Avra, Red Glass Ventures, Carpenter Capital, Audacious, PSP, Overmatch Ventures, XYZ, Haystack, Brave Capital.
Learn more about this fundraise:
bloomberg.com/news/articles/…
Sequoia Capital retweeted
We dug into the hand logs from the Bay Area regional. No single dominant strategy:
Winning bots showed up in every corner of the preflop map, tight and loose, passive and aggressive.
Same story with the card they chose to show.
Can’t share the full playbooks yet. Two more regionals to go, two new variants incoming 👀
150 teams showed up to compete in the first regional showdown of the No Limit tournament on Friday night:
- 240+ students
- 20+ universities
- 60 minutes to build
- Poker variant: Hold ‘em with three hole cards, post-flop discard one card, expose one to the table
- Pokerbots played 300M+ hands across six rounds
Lessons learned? 1) Bots believe what they see: show a card that pairs the board and opponents fold 72% of the time 2) Nobody bluffs the river 3) All-in bots do not win anymore.
5 winners are headed to the Finals in San Francisco this Sunday, 11/15.
Applications for the Pittsburgh regional close tonight at 11:59 PM ET.
Thanks to our partners @OpenAI @citsecurities.
150 teams showed up to compete in the first regional showdown of the No Limit tournament on Friday night:
- 240+ students
- 20+ universities
- 60 minutes to build
- Poker variant: Hold ‘em with three hole cards, post-flop discard one card, expose one to the table
- Pokerbots played 300M+ hands across six rounds
Lessons learned? 1) Bots believe what they see: show a card that pairs the board and opponents fold 72% of the time 2) Nobody bluffs the river 3) All-in bots do not win anymore.
5 winners are headed to the Finals in San Francisco this Sunday, 11/15.
Applications for the Pittsburgh regional close tonight at 11:59 PM ET.
Thanks to our partners @OpenAI @citsecurities.
Sequoia Capital retweeted
Continual learning is the true north for a lot of the progress in AI. It honors human experience by making it legible to AI, with alignment baked in. Ronak does an awesome job of breaking this down in our latest episode of Training Data.
Models have IQ but no experience.
Frontier models show up on day one with superhuman intelligence but no expertise on the thousands of tasks your company actually does. @trajectorylabs exists to close that gap through continual learning.
@rronak_ trained SWE-1 at Windsurf, then went to DeepMind. His cofounders @MichaelElabd and @QuantumArjun worked on Gemini Robotics and multimodal models for Vision Pro. All three hit the same wall: big labs train one model for everyone, but nobody trains one for you.
The conversation covers:
— why today’s agents are “pseudo continual learning”: we are in the paper road map era, not Google Maps
— “Group Size 1”, a GRPO variant that learns from one real-world rollout instead of replaying the same task eight times, so you don’t need a simulated environment
— how Harvey trained a model 6x cheaper than frontier that took an open-source baseline up by over 4x on its own legal benchmark
— Continual Grading, a meta-grader that turns rubrics on 100 traces into reward for 10,000
— continual learning vs. recursive self-improvement, and why only one keeps humans in the loop
— why continual learning will make product-building more like tending a garden of products
Chapters
0:00 – Introduction
0:45 – Pseudo continual learning: paper maps vs Google Maps
2:48 – Do we need new architectures? Rich Sutton, David Silver, and the missing layer
4:33 – From a $100M consulting model to a product: lessons from SWE-1 at Windsurf
7:36 – Harvey: 6x cheaper than frontier, 4% to 18% on its own benchmark
8:55 – Why "Trajectory": the unit of work is a task, not a token
10:13 – The primitives of continual learning: traces, replayable environments, reward
12:56 – Group Size 1 and the meta-grader: training without an environment or a labeled reward
16:16 – The platform: train, eval and deploy in one click, and who it's for
18:58 – Rogo and intelligence density
20:14 – What's next: curiosity, and why continual learning beats RSI on alignment
24:19 – Research and product under one roof, and the product builder as gardener
27:26 – Why work at Trajectory, and the optimistic case: from 1% to 99%
Sequoia Capital retweeted
Models have IQ but no experience.
Frontier models show up on day one with superhuman intelligence but no expertise on the thousands of tasks your company actually does. @trajectorylabs exists to close that gap through continual learning.
@rronak_ trained SWE-1 at Windsurf, then went to DeepMind. His cofounders @MichaelElabd and @QuantumArjun worked on Gemini Robotics and multimodal models for Vision Pro. All three hit the same wall: big labs train one model for everyone, but nobody trains one for you.
The conversation covers:
— why today’s agents are “pseudo continual learning”: we are in the paper road map era, not Google Maps
— “Group Size 1”, a GRPO variant that learns from one real-world rollout instead of replaying the same task eight times, so you don’t need a simulated environment
— how Harvey trained a model 6x cheaper than frontier that took an open-source baseline up by over 4x on its own legal benchmark
— Continual Grading, a meta-grader that turns rubrics on 100 traces into reward for 10,000
— continual learning vs. recursive self-improvement, and why only one keeps humans in the loop
— why continual learning will make product-building more like tending a garden of products
Chapters
0:00 – Introduction
0:45 – Pseudo continual learning: paper maps vs Google Maps
2:48 – Do we need new architectures? Rich Sutton, David Silver, and the missing layer
4:33 – From a $100M consulting model to a product: lessons from SWE-1 at Windsurf
7:36 – Harvey: 6x cheaper than frontier, 4% to 18% on its own benchmark
8:55 – Why "Trajectory": the unit of work is a task, not a token
10:13 – The primitives of continual learning: traces, replayable environments, reward
12:56 – Group Size 1 and the meta-grader: training without an environment or a labeled reward
16:16 – The platform: train, eval and deploy in one click, and who it's for
18:58 – Rogo and intelligence density
20:14 – What's next: curiosity, and why continual learning beats RSI on alignment
24:19 – Research and product under one roof, and the product builder as gardener
27:26 – Why work at Trajectory, and the optimistic case: from 1% to 99%
Sequoia Capital retweeted
@PlayAstrocade minutes per user per day metric is getting to Instagram level on a base of millions of users. This is attracting more creators, and there is a real flywheel here. Now we're putting fuel on the 🔥with the Astrocade Creator Competition. Let's go!!
Sequoia Capital retweeted
Today we release Ripple-65B Preview, our 65B dense LLM we pretrained from scratch.
Ripple sets a new state-of-the-art for U.S. open-weight models in agentic coding tasks, surpassing significantly larger models including Muse Spark Open-816B, Beam-501B, Inkling-975B, and Nemotron 3 Ultra-550B.
Learn more: ricursive.com/ripple