正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

We've raised $7M to help companies build AI agents that actually learn and work. Osmosis (YC W25) is a platform for companies to fine-tune models that outperform foundation models with reinforcement learning. Better, faster, and cheaper.

Kasey Zhang

2,733 subscribers

1,217,447 次观看 • 11 个月前 •via X (Twitter)

科学技术新闻政治财经

Anya Rossi• Live Now

Private livecam show

50 条评论

Kasey Zhang 的头像

Kasey Zhang11 个月前

Our seed round was co-led by @CRV and @AudaciousHQ. We’re honored to be supported by angels like @paulg, @rauchg, @MishaLaskin, @bernhardsson, @bryantchou, @saradu, @m_franceschetti, @waseem_s, Troy Demmer, @shaneguML, @andrewztan, @jackyhuang, and many others.

Kasey Zhang 的头像

Kasey Zhang11 个月前

We’re hosting RL IRL in 10 days with @ycombinator, @shaneguML, @reductoai, @greptile, @encord_team, and @hud_evals - it’ll be a great opportunity to meet others working on applied reinforcement learning!

Kasey Zhang 的头像

Kasey Zhang11 个月前

If you want to learn more about Osmosis, we'd love to chat:

Guillermo Rauch 的头像

Guillermo Rauch11 个月前

@Osmosis_AI congrats, been enjoying working with you guys

Kasey Zhang 的头像

Kasey Zhang11 个月前

@Osmosis_AI Likewise! Excited to be hill climbing with y'all 📈

parth 的头像

parth11 个月前

@jackyhuang @Osmosis_AI Congrats on the launch! I want to see a Cambrian explosion of SOTA open models, RL’d on specific use cases.

Kasey Zhang 的头像

Kasey Zhang11 个月前

@jackyhuang @Osmosis_AI Admin he's hill climbing sideways

david 的头像

david11 个月前

@Osmosis_AI this is a nice vid man

Kasey Zhang 的头像

Kasey Zhang11 个月前

@Osmosis_AI Had to get something nice after you guys raised the bar 🫡

helen ⁎⁺˳✧༚ 的头像

helen ⁎⁺˳✧༚11 个月前

@Osmosis_AI GOAT, OF THE MOUNTAIN VARIETY !!! FINDING FOOTHOLDS WHERE OTHERS FALL

Kasey Zhang 的头像

Kasey Zhang11 个月前

@Osmosis_AI Of the mountain variety, as opposed to the pog variety

helen ⁎⁺˳✧༚ 的头像

helen ⁎⁺˳✧༚11 个月前

@Osmosis_AI that’s a shiny variant

Jonathan Chang 的头像

Jonathan Chang11 个月前

@Osmosis_AI Huge congrats!

zehra ✨ 的头像

zehra ✨11 个月前

@Osmosis_AI !!!!!!!!!!! sick

Kasey Zhang 的头像

Kasey Zhang11 个月前

@Osmosis_AI Thanks Zehra!! The @loreobsessed video was insaneee

David Zhi LuoZhang 的头像

David Zhi LuoZhang11 个月前

@Osmosis_AI Congrats Kasey!!

Misha Laskin 的头像

Misha Laskin11 个月前

@Osmosis_AI Congrats Kasey!

Kasey Zhang 的头像

Kasey Zhang11 个月前

@Osmosis_AI Thanks so much Misha!

Muzzammil Zaveri (MZ) 的头像

Muzzammil Zaveri (MZ)11 个月前

@Osmosis_AI Congrats @_WEEXIAO @viewerisland @Osmosis_AI! Excited to be a part of your journey 🙌

jacky (: 的头像

jacky (:11 个月前

@Osmosis_AI GOATTTTTT

Pranav Mital 的头像

Pranav Mital11 个月前

@Osmosis_AI This team doesn’t play around. Literally climbing hills (i)RL

Kasey Zhang 的头像

Kasey Zhang11 个月前

@Osmosis_AI We learn from experience...

Michael 的头像

Michael11 个月前

@Osmosis_AI this is so cool!!

Kasey Zhang 的头像

Kasey Zhang11 个月前

@Osmosis_AI Appreciate it Michael!!

Michael Zhu 的头像

Michael Zhu11 个月前

@Osmosis_AI LFG congrats!!

jack mcclelland 的头像

jack mcclelland11 个月前

@Osmosis_AI congrats on the launch @_WEEXIAO

adel 🌟 的头像

adel 🌟11 个月前

@Osmosis_AI wohoo!!! go kasey

shreyansh🫧 的头像

shreyansh🫧11 个月前

@Osmosis_AI Now this is some interesting product

andrew pignanelli 的头像

andrew pignanelli11 个月前

@Osmosis_AI RL so hot these days

Eli Brown 的头像

Eli Brown11 个月前

@Osmosis_AI Congrats!!!

Kasey Zhang 的头像

Kasey Zhang11 个月前

@Osmosis_AI Thanks Eli!! Really appreciate your support and guidance thus far :)

Ryan 的头像

Ryan11 个月前

@Osmosis_AI 🐐

Shailee Sankhala 的头像

Shailee Sankhala11 个月前

@Osmosis_AI wow wish i had rl for ai back in the day smh

Sriman Gaddam 的头像

Sriman Gaddam11 个月前

@Osmosis_AI Congrats on the launch 🚀

Tom Greenwald 的头像

Tom Greenwald11 个月前

@Osmosis_AI Congrats man!

Rohin Arya 的头像

Rohin Arya11 个月前

@Osmosis_AI Congrats Kasey and team!

Adish Jain ☕️ 的头像

Adish Jain ☕️11 个月前

@Osmosis_AI Congrats team!!!

Ryan D’Onofrio 的头像

Ryan D’Onofrio11 个月前

@Osmosis_AI Congrats team! Excited to learn more next week and see how it might enable new capabilities in the healthcare space!

Kasey Zhang 的头像

Kasey Zhang11 个月前

@Osmosis_AI Excited to see you at RL IRL!

Brenda 的头像

Brenda11 个月前

@Osmosis_AI congrats!! 🥳

anirudh 的头像

anirudh11 个月前

@Osmosis_AI huskies on top red black one pack lfg kasey

Kasey Zhang 的头像

Kasey Zhang11 个月前

@Osmosis_AI Alexa play Stacy's Mom

Augustin 🐓 的头像

Augustin 🐓11 个月前

@Osmosis_AI looks nice!

Kasey Zhang 的头像

Kasey Zhang11 个月前

@Osmosis_AI Thanks Augustin!

Adhityaa 的头像

Adhityaa11 个月前

@Osmosis_AI 🐐

Aryan Jain 的头像

Aryan Jain11 个月前

@Osmosis_AI Congrats guys!

Shrey Pandya 的头像

Shrey Pandya11 个月前

@Osmosis_AI Congrats!!

Hadi 的头像

Hadi11 个月前

@Osmosis_AI congrats! and amazing video😃

Charlie Deane 的头像

Charlie Deane11 个月前

@Osmosis_AI LETS GOOOO

Manya Adams 的头像

Manya Adams11 个月前

@Osmosis_AI wow...I wonder when I'll get to my first million

相关视频

Halluminate is building reinforcement learning environments and benchmarks to help AI models do knowledge work beyond coding, starting with finance. With a team of fewer than 10 people, the company works with four of the top five closed-source US AI labs and recently raised a $30M Series A. In this fireside, co-founders Jerry Wu and wyatt marshall sit down with YC's Jon Xu to share how testing browser agents led them to building the environments models now use to practice tasks like financial modeling. They explain why better verification, not just more training tasks, is critical to making models smarter, and how that work could eventually lead to simulated companies where teams of AI agents learn to work together. 00:00 — Halluminate’s Pivot From Evals to Training Environments 03:48 — Teaching AI to Do Financial Work 07:13 — Why Verification Matters More Than Volume 10:13 — From Finance to Simulated Companies 13:30 — AI Safety and Building the Team

Halluminate is building reinforcement learning environments and benchmarks to help AI models do knowledge work beyond coding, starting with finance. With a team of fewer than 10 people, the company works with four of the top five closed-source US AI labs and recently raised a $30M Series A. In this fireside, co-founders Jerry Wu and wyatt marshall sit down with YC's Jon Xu to share how testing browser agents led them to building the environments models now use to practice tasks like financial modeling. They explain why better verification, not just more training tasks, is critical to making models smarter, and how that work could eventually lead to simulated companies where teams of AI agents learn to work together. 00:00 — Halluminate’s Pivot From Evals to Training Environments 03:48 — Teaching AI to Do Financial Work 07:13 — Why Verification Matters More Than Volume 10:13 — From Finance to Simulated Companies 13:30 — AI Safety and Building the Team

Y Combinator

258,252 次观看 • 4 天前

Despite constant improvements, AI agents still struggle with complex, multi-step tasks. @Gulp_AI (YC W25) solves this with Osmosis: a framework that helps AI agents learn (similar to DeepSeek R1) in real time. Congrats on the launch Kasey Zhang + Andy!

Despite constant improvements, AI agents still struggle with complex, multi-step tasks. @Gulp_AI (YC W25) solves this with Osmosis: a framework that helps AI agents learn (similar to DeepSeek R1) in real time. Congrats on the launch Kasey Zhang + Andy!

Y Combinator

78,484 次观看 • 1 年前

.Decagon co-founder and CTO Ashwin Sreenivas says the choice between expensive frontier models and cheaper, less capable ones is a false trade-off: "Even if you have a 'dumber model,' you can get it to higher performance on that specific task." "When we fine-tune smaller, dumber models, it's that they're just not as general purpose, but on the specific task we want them to do, they actually outperform the large, smart, state-of-the-art models." "So we end up getting all three things. It is better at the task, it is cheaper, and it is faster." Ashwin Sreenivas

.Decagon co-founder and CTO Ashwin Sreenivas says the choice between expensive frontier models and cheaper, less capable ones is a false trade-off: "Even if you have a 'dumber model,' you can get it to higher performance on that specific task." "When we fine-tune smaller, dumber models, it's that they're just not as general purpose, but on the specific task we want them to do, they actually outperform the large, smart, state-of-the-art models." "So we end up getting all three things. It is better at the task, it is cheaper, and it is faster." Ashwin Sreenivas

a16z

36,067 次观看 • 2 个月前

Introducing Billow (YC S26) We're killing Deloitte. Accounting is a $700B industry, but 80% of companies still do it by hand. Billow is an AI-native accounting firm that replaces humans with agents. We connect to your existing tools (NetSuite, QuickBooks, and 80+ other systems) and provide faster, cheaper accounting services. There is ZERO software for customers to use. Just integrate, and Billow closes your books. We're live with enterprise and public companies representing over $1B in value.

Introducing Billow (YC S26) We're killing Deloitte. Accounting is a $700B industry, but 80% of companies still do it by hand. Billow is an AI-native accounting firm that replaces humans with agents. We connect to your existing tools (NetSuite, QuickBooks, and 80+ other systems) and provide faster, cheaper accounting services. There is ZERO software for customers to use. Just integrate, and Billow closes your books. We're live with enterprise and public companies representing over $1B in value.

Joanathan McIntosh

402,790 次观看 • 1 个月前

Plexe AI (Plexe AI) is building an open-source agent to build ML models from a prompt. Their agent is already helping companies go from idea to deployed models 10x faster. Congrats on the launch, Vaibhav Dubey and Marcello De Bernardi!

Plexe AI (Plexe AI) is building an open-source agent to build ML models from a prompt. Their agent is already helping companies go from idea to deployed models 10x faster. Congrats on the launch, Vaibhav Dubey and Marcello De Bernardi!

Y Combinator

55,896 次观看 • 1 年前

Specialized AI agents help enterprises turn AI into systems built for their own workflows. NVIDIA Agent Toolkit brings together open Nemotron models, tools, skills and secure runtime support to help teams build agents tuned for domain-specific work. Learn more:

Specialized AI agents help enterprises turn AI into systems built for their own workflows. NVIDIA Agent Toolkit brings together open Nemotron models, tools, skills and secure runtime support to help teams build agents tuned for domain-specific work. Learn more:

NVIDIA

81,019 次观看 • 3 个月前

⚡Faster models mean smarter AI. ⚡ With Nemotron's open-source technologies, we're enabling organizations to create AI agents that get more done in less time. Big or small, we are committed to helping companies all around the world build & deploy AI for their own needs. See how:

⚡Faster models mean smarter AI. ⚡ With Nemotron's open-source technologies, we're enabling organizations to create AI agents that get more done in less time. Big or small, we are committed to helping companies all around the world build & deploy AI for their own needs. See how:

NVIDIA

35,710 次观看 • 1 年前

The founder of a $4B inference company says that if you're building agents, foundational models could become your IP. According to Lin Qiao, 90% of the world's data is still private and locked inside applications and enterprises. Foundation models are trained on public internet + labeling company data, which is barely 10% of all the data. And this is why your application and foundation models are misaligned by definition. Companies building agents don't treat models as APIs. They opt for a product-model co-design. - Models continuously learn from your private data - Pick up domain-specific intelligence - Run faster - Cost less - Scale to millions of users

The founder of a $4B inference company says that if you're building agents, foundational models could become your IP. According to Lin Qiao, 90% of the world's data is still private and locked inside applications and enterprises. Foundation models are trained on public internet + labeling company data, which is barely 10% of all the data. And this is why your application and foundation models are misaligned by definition. Companies building agents don't treat models as APIs. They opt for a product-model co-design. - Models continuously learn from your private data - Pick up domain-specific intelligence - Run faster - Cost less - Scale to millions of users

Ivan Burazin

55,344 次观看 • 5 个月前

YC W24 companies like Sync Labs have instant access to GPUs for fine tuning and training foundation models, and get $500K in credits so they don't even have to touch their balance sheet to show results like this below. You don't have to start with $100M on day one in AI

YC W24 companies like Sync Labs have instant access to GPUs for fine tuning and training foundation models, and get $500K in credits so they don't even have to touch their balance sheet to show results like this below. You don't have to start with $100M on day one in AI

Garry Tan

76,971 次观看 • 2 年前

INSANE how most AI agents still send every task to Claude Code. Jev routes the easy work to cheaper workers, saves the expensive reasoning for what actually matters, and makes the whole system literally faster. This is how you build agents that scale. Follow CyrilXBT for more AI updates.

INSANE how most AI agents still send every task to Claude Code. Jev routes the easy work to cheaper workers, saves the expensive reasoning for what actually matters, and makes the whole system literally faster. This is how you build agents that scale. Follow CyrilXBT for more AI updates.

CyrilXBT

17,849 次观看 • 11 天前

1/ Announcing the new Lindy: the first platform letting you build a team of AI employees that work together to perform any task — 100x better, faster and cheaper than humans would. Watch how I build a Lindy to track competitors in 30s:

1/ Announcing the new Lindy: the first platform letting you build a team of AI employees that work together to perform any task — 100x better, faster and cheaper than humans would. Watch how I build a Lindy to track competitors in 30s:

Flo Crivello

1,085,030 次观看 • 2 年前

The best way to learn AI is to build with agents. To help with that, we've launched hands-on labs and a new series on Agentic Engineering. First topic: Agent Skills. Next in the pipeline: planning, context engineering, multi-agent systems, long-running agents,.. Go build!

The best way to learn AI is to build with agents. To help with that, we've launched hands-on labs and a new series on Agentic Engineering. First topic: Agent Skills. Next in the pipeline: planning, context engineering, multi-agent systems, long-running agents,.. Go build!

elvis

31,802 次观看 • 4 个月前

AI companies are hijacking you to build their models Veri is your Personal Model Company meet your personal agents powered by a model you own.

AI companies are hijacking you to build their models Veri is your Personal Model Company meet your personal agents powered by a model you own.

Veri

75,395 次观看 • 10 天前

Jack Dorsey says closed-source AI models are able to manipulate users to align with the financial incentives of the companies that build them

Jack Dorsey says closed-source AI models are able to manipulate users to align with the financial incentives of the companies that build them

Tsarathustra

155,965 次观看 • 2 年前

We got into Y Combinator 3 months ago with zero revenue and today Fabraix (YC S26) is at $1.04M annualised run rate. Traditional cybersecurity was built for software that behaves within defined boundaries. AI agents can reason, use tools, take actions, and find ways around those boundaries. The cost asymmetry we're seeing in irl warfare is also now a software problem. So we built autonomous red-teaming agents that find real vulnerabilities in AI agents that work 24/7, powered by our own state-of-the-art offensive models that outperform frontier models. AI agents are automating all knowledge work. Fabraix (YC S26) is the adversarial testing infrastructure that will secure every production agent.

We got into Y Combinator 3 months ago with zero revenue and today Fabraix (YC S26) is at $1.04M annualised run rate. Traditional cybersecurity was built for software that behaves within defined boundaries. AI agents can reason, use tools, take actions, and find ways around those boundaries. The cost asymmetry we're seeing in irl warfare is also now a software problem. So we built autonomous red-teaming agents that find real vulnerabilities in AI agents that work 24/7, powered by our own state-of-the-art offensive models that outperform frontier models. AI agents are automating all knowledge work. Fabraix (YC S26) is the adversarial testing infrastructure that will secure every production agent.

Zach AL

45,619 次观看 • 24 天前

NVIDIA AQCURIES HUGGING FACE FOR $12.9B. Jensen on the value of open source models: "I recommended that people use closed models as much as they can because they are developing rapidly. We use Claude, Codex, Perplexity, and Cursor at Nvidia. However, there are so many companies that have domain expertise, companies that have sovereignty reasons, they do their own research, those companies need to have control and ownership over their own intellectual proprietary. Open sourced models have become sufficiently good to allow companies to build their own AI."

NVIDIA AQCURIES HUGGING FACE FOR $12.9B. Jensen on the value of open source models: "I recommended that people use closed models as much as they can because they are developing rapidly. We use Claude, Codex, Perplexity, and Cursor at Nvidia. However, there are so many companies that have domain expertise, companies that have sovereignty reasons, they do their own research, those companies need to have control and ownership over their own intellectual proprietary. Open sourced models have become sufficiently good to allow companies to build their own AI."

amit

122,906 次观看 • 1 个月前

Frontier models that use reasoning to "think" during inference are generating 5X more AI tokens per year. ✨ "Inference is now a thinking process. And in order to teach AI how to think, reinforcement learning and very significant computation was introduced into post-training," said NVIDIA CEO Jensen Huang in a recent keynote at #CES2026. Reinforcement learning is increasing computation demands across all AI scaling laws: pre-training, post-training, and test-time scaling. Learn more about reinforcement learning and AI scaling ➡️

Frontier models that use reasoning to "think" during inference are generating 5X more AI tokens per year. ✨ "Inference is now a thinking process. And in order to teach AI how to think, reinforcement learning and very significant computation was introduced into post-training," said NVIDIA CEO Jensen Huang in a recent keynote at #CES2026. Reinforcement learning is increasing computation demands across all AI scaling laws: pre-training, post-training, and test-time scaling. Learn more about reinforcement learning and AI scaling ➡️

NVIDIA AI Infrastructure

13,329 次观看 • 8 个月前

Today, we’re announcing Parahelp to the world and that we've raised $21.2M. In the past year, Parahelp has become the go-to AI support agent for the world’s fastest-moving software companies, and we're proud to say that 0% of the companies we've onboarded have churned. We’re hiring exceptional engineers, so if you’re excited to build on the frontier of applied AI for customer support, please reach out!

Today, we’re announcing Parahelp to the world and that we've raised $21.2M. In the past year, Parahelp has become the go-to AI support agent for the world’s fastest-moving software companies, and we're proud to say that 0% of the companies we've onboarded have churned. We’re hiring exceptional engineers, so if you’re excited to build on the frontier of applied AI for customer support, please reach out!

Anker

358,375 次观看 • 1 年前

Today, we're launching Belvedir We're making private AI models cheap and easy so the world can have trillions of them. Private AI models are cheaper, better, faster, and protect your data from frontier labs, but they're still to expensive and difficult to make for startups and consumers. Belvedir makes private AI models autonomously with less than 5 minutes of setup, so that private AI can finally be accessible to everyone

Today, we're launching Belvedir We're making private AI models cheap and easy so the world can have trillions of them. Private AI models are cheaper, better, faster, and protect your data from frontier labs, but they're still to expensive and difficult to make for startups and consumers. Belvedir makes private AI models autonomously with less than 5 minutes of setup, so that private AI can finally be accessible to everyone

Zachary Yu

89,695 次观看 • 1 个月前