Prompt Engineering for Generative AI: Future-Proof Inputs for Reliable AI Outputs

Somewhere in the past year, ChatGPT has gone from "cool, interesting, amusing" to a massively valuable work assistant capable of writing Python scripts, analysing data, and doing lots, lots, more. The key to this? The rapid evolution of OpenAI's GPT models and making my first forays into prompt engineering.If I thought I was getting good, though, this book took me reminded me that I'm just scratching the surface. Halfway through the first chapter I was already furiously scribbling notes in the margins for what I could do better with my prompt writing and by the end of the text I felt like I had gotten a very good grounding - not just in GPTs specifically but in the bigger picture of how these hugely powerful tools were trained and came to maturity.I imagine that few will argue with my assertion that there is lots of hyperbole and "noise" in the AI space right now which, as ever, makes it hard to pick out the signal from the noise. Which is precisely why I sought out an O'Reilly title and I'm very glad that I did.Thorough, excellent, and I hope that this edition will be the first of many. As this rapidly maturing field scales and matures I think that prompt engineering will be an essential discipline to master. Pick up this text to get a good foothold on things.

This is a solid book for understanding the art and science of working with LLMs and other generative AI models. I always struggled with getting the output I was looking for, and wasn't sure how best to "ask" the models for what I wanted. This book did a great job of laying out the strategies and practical guidance to craft the prompts. There were a lot of tips and tricks, but the overall understanding and framework around prompt engineering has been super useful.

I applaud the authors for putting forth a comprehensive introduction in a rapidly evolving space. I absorbed a lot, helpful as I was developing a prototype — using open source methods.And that’s where I was disappointed. The LLM and examples are highly adapted to OpenAI’s GPT-x and the bulky LangChain framework, something not obvious until you dig in to the book. Sure, this may be where newbie demand was when the authors began writing. But as the open source models and OpenAI alternatives gain speed (e.g. Llama 3.1, Groq, etc.) this book may quickly need an updated and expanded version to stay relevant.

I know this may seem superficial, but the fact that ORA has now included color printing to their zoo collection takes this to another level.

I was under the impression it would really help me with ChatGPT 3.5 or 4.0. It doesn't.

Hands down, the best book on prompt engineering and implementing LLMs. Really enjoyed Michael and James deep dive. Whether you're technical or not, this book is foundational to a deeper understanding in how to properly explore and implement LLMs.

If you don't know how to prompt AI models correctly, you're missing out on substantially better results. This book has literally everything I could have asked for, it's an awesome resource.

I live in the world of AI, yet already learned so much from the beginning of the book. The things I do daily, by rote, now have frameworks, logic, and explanations for why they are so. Looking forward to the rest of the book!