Vishal Satish

14 posts

Vishal Satish

Vishal Satish

@vsatish_

CTO @BalerionAI // Forward-deployed agents & RL for post-training // @Agentica_ Project // EECS @ UC Berkeley @berkeley_ai

SF/NYC Katılım Ağustos 2022
34 Takip Edilen33 Takipçiler
Sabitlenmiş Tweet
Vishal Satish
Vishal Satish@vsatish_·
Excited to announce our $6M seed round led by @kleinerperkins to build the next generation of agentic systems that thrive in messy real-world environments: @BalerionAI is taking AI out of the lab and putting it in the hands of hundreds of real world lenders to realize the American dream of home ownership and bring mortgage lending back to its fundamentals: a relationship-driven business, not the costly operational gauntlet it has become. To this end we are building the agentic copilot for lending that helps lenders move loans across the finish line faster. We’re thrilled to be working with @josh_coyne, joined by @formation_vc, @BoxGroup, @thehousefund, and an all-star line-up of operators and investors across the financial services. And last but not least, we’ve assembled a world-class team to tackle this challenge: ex-operators and leaders who have built, scaled, and pushed the boundaries of AI across robotics, financial services, gaming, and some of the most exciting vertical tech companies out there. If you think that’s you, let’s chat. This is just the beginning, and we’re fired up for the journey ahead. Take a look: balerion.ai
Balerion AI@BalerionAI

Mortgage lending has lost its way. It’s one of the most manual, fragmented, and costly workflows in financial services for everyone involved. We’re building toward a self-driving mortgage: a world where lenders can spend less time pushing files and more time building trust with borrowers. Our approach is simple in principle, but powerful in execution: an agentic AI copilot that coordinates the full loan lifecycle, from origination through closing, removing bottlenecks and accelerating every step along the way. Today, we’re announcing Balerion AI and our $6M seed round led by @kleinerperkins , alongside @formation_vc , @BoxGroup , and an exceptional group of operators and investors to realize that vision. If you’re a lender who wants to move faster and operate smarter while handling more volume, or if you’re excited by the AI engineering and infrastructure behind a self-driving mortgage, we’d love to talk. axios.com/pro/fintech-de…

English
4
3
11
560
Vishal Satish retweetledi
Agentica Project
Agentica Project@Agentica_·
🚀 Introducing DeepSWE 🤖: our fully open-sourced, SOTA software engineering agent trained purely with RL on top of Qwen3-32B. DeepSWE achieves 59% on SWEBench-Verified with test-time scaling (and 42.2% Pass@1), topping the SWEBench leaderboard for open-weight models. 💪DeepSWE is trained with rLLM, our modular RL post-training framework for agents. rLLM makes it easy to build, train, and deploy RL-tuned agents on real-world workloads — from software engineering to web navigation and beyond. 🤗As always, we’re open-sourcing everything: not just the model, but the training code (rLLM), dataset (R2EGym), and training recipe for full reproducibility. 🔥Train DeepSWE yourself. Extend it. Build your own local agents. No secrets, no barriers. DeepSWE and rLLM mark our major shift: from training language reasoners to building language agents that can truly learn from experience. We believe the future of AI lies in experience-driven learning — and we’re here to democratize it. Welcome to the era of experience. 🌍 Links below: (1/n)
Agentica Project tweet media
English
16
75
369
73.1K
Vishal Satish retweetledi
Agentica Project
Agentica Project@Agentica_·
We're trending on @huggingface models today! 🔥 Huge thanks to our amazing community for your support. 🙏
Agentica Project tweet media
English
2
6
44
2.8K
Vishal Satish retweetledi
Sijun Tan
Sijun Tan@sijun_tan·
Hey @sama, we know you're planning to open-source your reasoning model—but we couldn’t wait. Introducing DeepCoder-14B-Preview: a fully open-source reasoning model that matches o1 and o3-mini on both coding and math. And yes, we’re releasing everything: model, data, code, and the full training recipe. We can't wait to try out your model and train on top of it, looking forward to your release!
Agentica Project@Agentica_

Introducing DeepCoder-14B-Preview - our fully open-sourced reasoning model reaching o1 and o3-mini level on coding and math. The best part is, we’re releasing everything: not just the model, but the dataset, code, and training recipe—so you can train it yourself!🔥 Links below:

English
23
140
1.6K
150.2K
Vishal Satish retweetledi
Michael Luo
Michael Luo@michaelzluo·
🚀 We introduce DeepCoder-14B-Preview, a fully open-sourced coding model that is on par with o3-mini and o1! 📷 We scaled our model with RL magic up to 32K context. It's performance scales to 64K context 🔥
Michael Luo tweet media
Agentica Project@Agentica_

Introducing DeepCoder-14B-Preview - our fully open-sourced reasoning model reaching o1 and o3-mini level on coding and math. The best part is, we’re releasing everything: not just the model, but the dataset, code, and training recipe—so you can train it yourself!🔥 Links below:

English
9
14
113
11.4K
Vishal Satish retweetledi
Agentica Project
Agentica Project@Agentica_·
Introducing DeepCoder-14B-Preview - our fully open-sourced reasoning model reaching o1 and o3-mini level on coding and math. The best part is, we’re releasing everything: not just the model, but the dataset, code, and training recipe—so you can train it yourself!🔥 Links below:
Agentica Project tweet media
English
23
188
858
242.7K
Vishal Satish retweetledi
Agentica Project
Agentica Project@Agentica_·
Introducing Autellix: An agentic AI system that accelerates agentic applications. Run Deep Researcher, Google Co-Scientist, OAI Operator, or any program 💻—@langchain, @pyautogen, @crewAIInc, or just Python 🐍—4-15x faster than vLLM or SGLang⚡. Paper: arxiv.org/abs/2502.13965
English
2
14
45
7.8K
Vishal Satish retweetledi
Brandon Trabucco
Brandon Trabucco@brandontrabucco·
With the success of LLM agents like OpenAI Operator, we are entering a new scaling era, but how do we train these agent models? We present InSTA, the largest training environment for LLM agents, containing live web navigation tasks for 150k diverse websites in multiple languages. Website - data-for-agents.github.io Environment - github.com/data-for-agent… 🧵Thread below. 1/6 #AgenticAI #LLMs #OpenAl
GIF
English
9
29
162
37.8K
Vishal Satish
Vishal Satish@vsatish_·
Introducing PRIME-1: The First Vertically-Integrated AI Foundation Model for Warehouse Robots → Leverages 150k hours of robot data → Deployed in production for 3D tasks → Pre-trained on 1T tokens → Exhibits neural scaling laws ambirobotics.com/media/ambi-rob… Technical blog post to come!
Vishal Satish tweet media
English
1
3
8
754