Model Releases
I'm super excited to finally announce that we, @towards_AI, are releasing our second book, AI Engineering for Production. Put 15 experts, gr…
I'm super excited to finally announce that we, @towards_AI, are releasing our second book, AI Engineering for Production. Put 15 experts, graduates, and industry, and 3 years of work, and that's what
I'm super excited to finally announce that we, @towards_AI, are releasing our second book, AI Engineering for Production. Put 15 experts, graduates, and industry, and 3 years of work, and that's what you get: "...the most comprehensive textbook to date on building LLM applications, and helps learners understand everything from fundamentals to the simple-to-advanced building blocks of constructing LLM applications [...] - all essential topics in an AI Engineer's toolkit.” - @jerryjliu0 , CEO of @llama_index This is what Jerry said of our first book, Building a LLMs for Production, published in 2024. Now we put the same writing knowledge, even more effort, and more expertise in building with clients into a second entirely new book for AI engineers. We also took our learnings for creating long-lasting resources, even in this fast-evolving field. I believe that this second book is incredibly better and more useful than the first one and obviously fully up to date with the current state of AI. One of the reasons I quit my PhD in AI was to build practical solutions that will actually help others in the real world and improve what exists. While I love the world of academia, when I first stepped into the startup world in 2020, it felt like I pretty much knew nothing all over again. I needed to understand the problems in the real-world application of AI and build actual solutions for it that were useful to the company or to our clients, not just research. It wasn't anymore about just testing new models, testing new approaches, but really finding what works and what is reliable. Grasping these challenges and finding what's useful is just the first step. For the how part of it, you need to get into the code, the architecture, the models, the APIs, the deployment, the deployment aspect, the evaluation, the trial and error, the building, the complex and wide varieties of frameworks. You don't have time to reinvent the wheel in a startup. You need to use what exists: what is best, what will save you time, and what will work. Fortunately, Claude Code, and Codex can do wonders, but you still need true expertise for making all these decisions. We've gathered everything we worked on and with in this 500-plus-page book, all about LLMs and how to implement them into applications. Right now, this means working with context engineering, LLM judges, caching optimization, routing, RAG, tools, retries, agents, relevant frameworks and tools like ChromaDB, Langchain, Whisper, Gemini, Claude, and way more of such amazing tools But we believe the book teaches concepts that will stay relevant for a long time, even as LLMs and all these tools change and get better. We will always have to reduce hallucinations and work with the intrinsic limitations of language models based on how they are trained, how they learn, and how they are built. The only skill required for the book is some Python (or programming) knowledge. And as I said, this is not a one-person effort. I really want to thank some of our amazing team at Towards AI who really helped achieve this, including my co-author @_LouiePeters for his amazing contributions, and our AI engineers and editors Fabio Chiuzano, @omar_solano1 , Rucha Bhide, @NLPiation, @samridhivaid, and others… Also, a big thanks to the dozens of reviewers (and hundreds who’ve helped with feedback from this community!!) who helped us make the book as good as it could be. “AI Engineering for Production: Making LLMs and agents reliable with prompting, RAG evaluation, and harnesses” will be available at the end of November 2026. You can sign up to the waitlist here to get the book first. https://www.louisbouchard.ai/book/?utm_source=x-lf&utm_medium=social&utm_campaign=book I'll also host a nice live kickoff event, answering anything about the book and sharing our current AI engineering stack as of the end of 2026. If you want to support us for free, just join the waitlist, let us know your thoughts, and give us feedback. I really hope to see you at the kickoff launch, and hope you enjoy the book once it's out.
Related
Source: Jerry Liu (X) | 2026-10-26