Machine Learning
London, England
Joined April 2020
- Tweets1.8K
- Following225
- Followers270
- Likes974
Pinned Tweet
A book full of recipes to solve common NLP & computer vision tasks:
bipinKrishnan.github.io/ml-r…
Discusses the following libraries:
- @huggingface transformers, datasets, accelerate, @Gradio
- @PyTorchLightnin
- @wandb
- pytorch image models(timm)
- albumentations
Bipin Krishnan retweeted
Finally summarized what I've been sending to friends in DMs View article
Bipin Krishnan retweeted
Ever wondered what it’s like to learn robot autonomy @Stanford? Now you can.
Today, we’re opening up the robot autonomy teaching materials that my collaborators and I have been developing over the past several years—for anyone, anywhere, completely free.
At the heart of this effort is Principles of Robot Autonomy (PoRA), a textbook inspired by the Stanford robot autonomy course series I’ve been teaching since 2017.
Robot autonomy has evolved from a research discipline into a technology deployed in the real world. Building capable autonomous systems requires more than understanding individual algorithms: it requires understanding how perception, state estimation, planning, control, and learning work together in a complete autonomy stack.
That’s what we set out to teach with PoRA: bringing classical robotics and modern AI together in one coherent framework, grounded in both theory and practice.
Coming soon in print from Cambridge University Press, Principles of Robot Autonomy is also freely available online and comes with a comprehensive collection of open educational resources:
📖 Textbook, arxiv.org/abs/2608.03496 — Foundations of robot autonomy, from classical methods to modern AI-based approaches.
🎥 Video lectures, porabook.com/lectures/ Lectures from my Stanford courses on robot autonomy and on optimal and learning-based control.
💻 Interactive exercises, porabook.com/exercises/ — Hands-on Python implementations and Jupyter notebooks.
📝 Teaching materials, porabook.com/instructors/ — Lecture slides and resources for instructors.
🌐 Explore all PoRA resources, porabook.com/ — Including Ask the Book, an AI study assistant integrated throughout the textbook to answer questions, explain derivations, and connect ideas across chapters.
📣 Read the Stanford announcement: news.stanford.edu/stories/20…
Our goal is simple: to make high-quality education in robot autonomy accessible to students, researchers, engineers, educators, and anyone curious about how autonomous robotic systems work.
I believe that sharing knowledge is fundamental to the mission of a university. Autonomous robotics—and Physical AI more broadly—are among the defining technological frontiers of our time. We want to help more people understand these technologies, build with them, and advance the field.
A huge thank you to my co-authors, @DanieleGammelli, Joseph Lorenzetti, @katielulula, and @GioeleZardini, and to everyone who helped bring this project to life.
Explore the resources, use them in your own learning or teaching, and please share them with anyone who might find them useful!
#Robotics #PhysicalAI #ArtificialIntelligence #AutonomousSystems #Education
@Stanford @StanfordEng @StanfordAILab @StanfordHAI
Bipin Krishnan retweeted
We will be teaching Agentic GPU Programming for MLSys as part of the CMU ML Systems course this upcoming spring at @SCSatCMU, touching on knowledge curation, compiler analysis, and feedback loops. We also curated an online book on the topic, check it out mlc.ai/agentic-gpu-programmi…
Bipin Krishnan retweeted
After this, people kept asking us how @modal is able to serve agent inference so well.
So we wrote it all down.
New blog: "How to serve trillions of tokens for trillion-parameter coding agents".
modal.com/blog/trillion-toke…
Bipin Krishnan retweeted
LLMs keep getting bigger. Compute budgets, unfortunately, do not.
This fall @Stanford, we (@AnayMehrotra @gvelegkas and Amin Saberi) are teaching MS&E 319: Efficient Generative Language Models.
The course asks a simple question: given a modeling goal and a limited computational budget, how should we choose the training objective, model architecture, and inference algorithm?
We’ll cover some of the main ideas, and occasionally surprising tricks, that make large language models more efficient, including efficient pre-training, mixture-of-experts architectures, attention and KV-cache compression, quantization, speculative decoding, LoRA, RLHF, DPO, and distillation.
And no, “just buy more GPUs” will not be the only answer.
We’ll try to post all the lecture materials and recordings as the course unfolds.
web.stanford.edu/class/msand…
Bipin Krishnan retweeted
zero-to-sglang new chapters are out! 📖
Want to read inference engine source code but don’t know where to start?
We’ve just added two chapters to zero-to-sglang, the open-source AI Infra course from SGLang × Datawhale:
📖 2.1 — mini-sglang: What an Inference Engine Looks Like
📖 2.2 — The Journey of a Request + Code Walkthrough
Find us on GitHub: datawhalechina/zero-to-sglang 🫵
#SGLang #LLMInference #AIInfra #OpenSource #LLM
🤖 Made with AI
Bipin Krishnan retweeted
Sharing my first of hopefully many research blog posts! jasperlu.com/blog/training-s…
This one is the kind of educational blog post I wish I'd had when I started with RL for LLMs.
I tried to make it as open as possible. Every rollout is browsable, the code is open source, and I walk through my entire thought process, from learning rate sweeps to reward shaping.
Bipin Krishnan retweeted
I made a list of the 50+ best articles and resources I’ve read and relied upon in the last few years on open models and open-source AI. Happy reading :)
Bipin Krishnan retweeted
how inference engines actually work - releasing full talk slides!
i cover everything in the lifetime of a request e2e:
> the inference engine
> kv + prefix caching
> continuous batching
> paged attention
> chunked prefill
> sampling
> agentic loops from inside the engine
Bipin Krishnan retweeted
We’re teaching a new course at Penn: CIS 6280 · World Models 🌍
cis.upenn.edu/~cis6280/
Every day, LLMs amaze me with one more thing they can solve. But I’ve always felt there’s more to learn than what humans have written down—from observing the world, interacting with it, and experiencing what happens next. To me, that’s what world models are about—and why I see them as one of AI’s next big bets.
However, I struggled to find a systematic course on them—so we are building one!
Topics will include: representation learning, generative models, simulation, model-based RL, video and 3D generation, world models for robotics, reasoning, and code-based world models.
Hands-on work will include:
- Building an environment
- Training a world model
- Learning a policy
and a final research project with leaderboard.
We’re already six lectures in. Slides, demos, and readings are public, and we’ll keep adding materials throughout the fall.
Big thanks to our TA team— @TongMutianTMT @hagsaeng_bag @EnxinSong @KeelyAi04 —for helping bring this course to life!
Bipin Krishnan retweeted
so many cool classes coming up in the fall quarter!
- Deep Learning Alchemy by @kothasuhas @tatsu_hashimoto @wen_kaiyue : deep-learning-alchemy.github…
- Engineering AI Agents by @Diyi_Yang @michaelryan207 @jyangballin : cs329z.stanford.edu/
- AI Agents by @gneubig @dan_fried : cmu-agents.com/
- World Models by @thoma_gu : cis.upenn.edu/~cis6280/
have fun learning my friends
Bipin Krishnan retweeted
🚀 Introducing DeepSeek-V4.1-Flash: smarter, faster, more efficient.
🔹 Introducing the smallest model in our new architecture family, with native visual understanding.
🔹 Designed for greater capability, faster inference, higher throughput, and scaling to larger models.
1/6
Bipin Krishnan retweeted
The new DeepSeek V4.1 Flash model is mindblowing - back on top of the open-source model leaderboard and extremely cheap.
It has a lot of very smart ways to be efficient and highly capable so I made a video of the forward pass to give you a view of what going on inside the model during inference.
Read more at huggingface.co/deepseek-ai/D…
And find the weights at huggingface.co/deepseek-ai/D…
Bipin Krishnan retweeted
First blog! Part 1 of writing speed-of-light GEMM kernels on Blackwell. This part covers basic GPU architecture, CuTeDSL fundamentals, loop tiling/TMA/TMEM, and swizzling.
lukehuang33.github.io/blog/b…
Bipin Krishnan retweeted
Announcing "DeepSeek-V3: from roofline to reality", a blog post series that digs into roofline analysis specifically on DSv3. We're kicking off this series with two posts! deepseek-v3.ezyang.com/index…
Bipin Krishnan retweeted
I present to you an Aug 2026 massively revised Machine Learning Engineering open book
github.com/stas00/ml-enginee…
If you prefer the pdf it's now 498 page long
github.com/stas00/ml-enginee…
The book went through a massive update to bring all the high end hw specs to being up-to-date, and updated many examples and studies to the recent torch / cuda and other critical in ML components.
Many bugs and inaccuracies have been fixed. If you read the book and find any remaining please kindly flag them to me.
Bipin Krishnan retweeted
Wrote up a blog on how the search engine of Papers with Code is built! 🔥
This involves a lot of @huggingface tooling
> PostGreSQL is used with pgvector
> the @Alibaba_Qwen 3 -Embedding-0.6B model is used
> hybrid search gets the best results
> embeddings are computed on an @nvidia L4 GPU powered by Hugging Face Jobs
> artifacts are stored in Buckets
> a live embedding endpoint is served on Hugging Face Inference Endpoints
> this also powers the "related papers" feature on each paper page
Learn more below!
huggingface.co/blog/pwc-sear…
Bipin Krishnan retweeted
Looks like they have released their technical report.
"Prime Agent: A Self-Improving RLM Harness"
arxiv.org/abs/2608.23552
The paper discusses a bit more about the design philosophy behind Prime Agent, particularly its role as an agent harness for long-horizon evaluation.
There are a few additional results in here as well. For example, there are more details about their Factorio results, where they did seven-day Sonnet 5 trajectory with 23.4 million output tokens, 24/196 technologies completed, 71% progress toward the next technology, 633 subagents, maximum of 7 simultaneously active agents
Bipin Krishnan retweeted
Getting ready to publish my complete guide to RL for LLMs tomorrow morning. Although the post contains many of my own thoughts / learnings, it is also a synthesis of so many great resources that have been published over the years:
- The RLHF Book (rlhfbook.com/) by @natolambert
- Reinforcement Learning by Richard S. Sutton and Andrew G. Barto
- Spinning Up in Deep RL (spinningup.openai.com/en/lat…) from OpenAI
- Build an LLM (sebastianraschka.com/llms-fr…) and Reasoning Model (sebastianraschka.com/reasoni…) from Scratch by @rasbt
- Various notes (joschu.net/blog/kl-approx.ht…) and papers (TRPO, PPO, etc.) from John Schulman
- Policy Gradient Algorithms (lilianweng.github.io/posts/2…) by Lilian Weng
- A Vision Researcher’s Guide to RL (yugeten.github.io/posts/2025…) by @YugeTen
- From REINFORCE to Dr. GRPO (lancelqf.github.io/note/llm_…) by @qingfeng_lan
- Async GRPO in the Wild (yumoxu.notion.site/async-grp…) by @yumo_xu
- Open RL infrastructure like TRL (huggingface.co/docs/trl/en/i…) and OpenInstruct (github.com/allenai/open-inst…)
I highly recommend reading all of them. They’ve truly helped me to learn so much.
Bipin Krishnan retweeted
I'm not 100% happy with this, but I am moving on and I think publishing is better than not publishing. So, with Fable and Sol: the interactive "How to Parallelize a Transformer for Training" ezyang.github.io/interactive… , based on the hit classic jax-ml.github.io/scaling-boo…
