BURKOV (@burkov) header
BURKOV (@burkov) profile picture

Books: https://xitter.cf/t.co/0EmPM3De9B & https://xitter.cf/t.co/45NGbbXIzC App: https://xitter.cf/t.co/n2jvMtYhVm PhD in AI, author of đź“– The Hundred-Page LMs Book & The Hundred-Page ML Book

Québec, Canada
Joined June 2009
Photo by BURKOV (@burkov)Photo by BURKOV (@burkov)Photo by BURKOV (@burkov)Photo by BURKOV (@burkov)Photo by BURKOV (@burkov)Photo by BURKOV (@burkov)Photo by BURKOV (@burkov)Photo by BURKOV (@burkov)Photo by BURKOV (@burkov)Photo by BURKOV (@burkov)
Pinned Tweet
As you know, last year I resigned from my full-time job and become a professional technical book writer. I have plans for The Hundred-Page Books about reinforcement learning, computer vision, diffusion models, and more. Not having a full-time job has put a significant strain on my savings, so if you love my books and are waiting for more, I would appreciate your continuous support. I have finalized the offer for those who financially support me. Here's what I offer: - As a Cool Supporter 10 🚀 ($10/month), you will get early access to new book chapters, exclusive updates on what to really care about in AI and why (only the important stuff, no hype), and a free e-book copy of all new books. - As a Cool Supporter 30 🚀 ($30/month), you will get everything a Cool Supporter 10 🚀 gets, plus a signed hard copy of all new books. The names of all Cool Supporters 🚀 will be mentioned in the Acknowledgment section of the new books as important contributors. Your name, printed in the book, will remain in history forever! If you decide to become a Cool Supporter 🚀, you can choose from several platforms. Here's the list: linktr.ee/burkov If you choose Substack, you will also get a subscription to a weekly AI newsletter without ads.
13
6
87
45,453
Just in case you didn't know: almost everything you send to OpenRouter is read by someone unless proven otherwise. Maybe large companies like OpenAI or Google don't do that; at least they have some internal access controls and might have whistleblowers if they don't respect their own obligations. But when you send a request to an open-weight model served by someone from their basement, I’d give it a 97.7% chance that your prompt is saved and then analyzed by the guy from the basement.
3
2
1
14
1,052
A free online book: Demystifying Artificial Intelligence by Emmanuel Gillain. Demystifying Artificial Intelligence is an introductory, practice-oriented textbook designed especially to help business practitioners understand the main techniques behind AI applications. It pairs accessible explanations with formal foundations and industry cases, giving readers a framework for judging what different AI methods do, where they are useful, how they relate, and what constraints shape their use. Read the book online on @ChapterPal: chapterpal.com/book/0e9c7329… The book comes in two versions: the original, as written by the authors (2024), and an edition updated for 2026 by ChapterPal's team. Use the built-in AI tutor when you feel stuck. Solve AI-generated quizzes to test your understanding.
1
1
19
1,015
There were times when you could be cool
20
1,703
A 1939 GM Futurliner
1
12
1,105
A free online book: Statistical Foundations of Reinforcement Learning: A Non-Asymptotic Perspective by Yuxin Chen, Yuejie Chi, Jianqing Fan, Gen Li, Yuting Wei, and Yuling Yan (2026). The book develops a rigorous account of how much data reinforcement-learning algorithms need to estimate values and learn effective policies. Centered on finite-state, finite-action Markov decision processes, the book studies finite-sample and finite-time performance, including settings where data are scarce. Its mathematical treatment draws on probability, statistics, information theory, and statistical learning theory, and presents largely self-contained proofs with intermediate arguments intended to build intuition. Readers interested in the theory and analysis of reinforcement learning will find a sustained comparison of algorithms’ guarantees and the fundamental limits those guarantees are measured against. The book is technical, so read it after you read my thedrlbook.com/ which is technical too, but in a manageable way. Read the book online on @ChapterPal: chapterpal.com/ebook/0cc5cf3…
1
27
1
171
4,075
Don't worry, keep being bullish; they’ll soon prove another theorem the world was just fine without! Next time on 500,000 pages!
4
2
34
1,526
Replying to @burkov
@burkov my bull is an AI skeptic but I got him interested in your latest book when I told him the TLDR is “chase girls”
1
2
5
365
A free online book (course notes): Foundations of Deep Learning by Chinmay Hegde. These graduate-level course notes develop a theoretical account of deep learning, using precise questions about neural networks to connect practical training behavior with ideas from statistical learning, approximation theory, and optimization. They organize the subject around three linked challenges: whether a network can represent a desired predictor, whether training can find a suitable model, and whether that model generalizes to unseen data. The notes establish this framework early and return to it as they move from representation to optimization and generalization. Read the book online on @ChapterPal: chapterpal.com/book/fac67658… When reading on ChapterPal, use the built-in AI tutor when you feel stuck. Solve AI-generated quizzes to test and solidify your understanding. Do exercises with the AI tutor, which guides you instead of solving them for you.
2
31
1
203
4,467
A free online book (course notes): Geometry Processing Research in Python by Oded Stein (2024). These course notes offer a practical starting point for newcomers to geometry processing research in Python, especially readers already familiar with Python and NumPy. Using Gpytoolbox for geometry operations and Polyscope for interactive visualization, they teach a workflow for loading, inspecting, modifying, and presenting meshes, then building small research methods from library functions. Read the course notes online on @ChapterPal: chapterpal.com/book/c4d335fa… Use the built-in AI tutor when you feel stuck. Solve AI-generated quizzes to test and solidify your understanding. Do exercises with the AI tutor that guides you instead of solving for you.
12
2
69
2,189
BURKOV retweeted
The Hundred-Page Machine Learning Book (PDF + EPUB + extra PDF formats) by Andriy Burkov is on sale on Leanpub! Its suggested price is $40.00; get it for $14.00 with this coupon: leanpub.com/theMLbook/c/Lean… @burkov #data_science #computer_science
1
3
484
A free online textbook: Parallel Programming for Science and Engineering by Victor Eijkhout. The textbook teaches how to design, implement, and assess programs that use multiple processors for scientific and engineering work. It is suited to readers working in applied mathematics, engineering, and the physical sciences; its examples and projects assume familiarity with programming, while individual chapters introduce the basic ideas and practical setup of each programming system. The book’s guiding approach is to build an accurate mental model of parallel execution and connect programming choices to data distribution, communication, correctness, and performance. Read the book online on @ChapterPal: chapterpal.com/book/90d8650c… Use the built-in AI tutor when you feel stuck. Solve AI-generated quizzes to test and solidify your understanding. Do exercises with the AI tutor that guides you instead of solving for you.
21
2
139
3,396
I don't know how you folks are capable of using GPT 6.2 Sol for coding. It's as dumb as Luna. Maybe dumber.
4
3
1
33
3,896
How does the suggested follow-up prompt in Codex know that, in the previous iteration, Codex fucked up and asks it to revert the fucked-up code, but the Codex agent doesn't know that and delivers the fuckup as a fix?
1
8
1,884
ChapterPal is becoming the destination for AI-assisted self-education on technical topics. If you haven't tried reading texbooks with its innovative reading app, you should. Dozens of users have told me they can't read in a traditional reader anymore. ChapterPal is the only place on the web where books are kept up to date. The author may have forgotten about their book and no longer care. We care.
2
3
1
29
1,868
A new issue of my weekly AI newsletter is out: Here be monsters: What do you still need to know when agents do the work? There’s no such thing as society in the objective function Opinion: “AI Promised Less Work. I Have Never Worked More.” When do causal ML models beat LM / GLM / GBM? Beyond Top-K: Building search-confidence guardrails for agentic AI with OpenSearch [Tool] DeepInfer: High-performance, large-scale inference framework for DeepSeek models [Google] llm-d 1.0: Open distributed inference for agents on Google Cloud TPUs and GPUs [Book] The Hundred-Page Deep Reinforcement Learning Book (2026) by Andriy Burkov [Research] A Circuit for In-Context Learning True Positive Weekly #186 aiweekly.substack.com/p/true…
3
3
9
1,781
A free online book: Operating System Fundamentals by Umar Khokhar and Binh Tran (2023). An operating system is the software layer that connects users and applications with computer hardware, allocates shared resources, and provides a platform for programs to run. The book develops this view from basic computer organization and OS goals into the mechanisms that make systems work, then applies those ideas to contemporary operating systems and cloud computing. The book is suited to readers seeking a structured introduction to operating-system concepts, including the terminology and models used to reason about system behavior. Read the book online on @ChapterPal: chapterpal.com/book/9231798a… Use the built-in AI tutor when you feel stuck. Solve AI-generated quizzes to test your understanding.
2
48
1
346
7,227
And I'm sad that you still have this boyfriend. Anyone who doesn't instantly feel like vomiting when they see slop in personal communication is seriously ill.
2
1
17
1,931