Building Applications with AI Agents: the UX chapter nobody else writes I read it half-published, with the measurement and production chapters still unreleased. Even so it delivered the hardest thing to find: how you present, to a person, a system that is sometimes wrong. AI 12 nov 2025 Books
Artificial Intelligence: A Modern Approach: the book that defines the word everyone misuses A thousand pages of reference. You do not read it straight through and you do not need to; chapter one alone dismantles half the sales conversations about agents. AI 31 oct 2025 Books
Artificial Intelligence in Finance: the best manual on lying to yourself with historical data Machine learning applied to markets. I neither trade nor advise on investments; I read it for the part nobody quotes, which is why a test on the past looks good and reality does not. AI Strategy 27 oct 2025 Books
Applied Machine Learning and AI for Engineers: a reminder that not everything is solved by conversation Regression, classification, networks, vision and text, explained for people who write code and did not come from statistics. Part of it aged badly and part of it is exactly what today's conversations are missing. AI 18 oct 2025 Books
An Illustrated Guide to AI Agents: the one I hand to whoever has to decide Drawings that genuinely explain. Memory, tools, planning, evaluation: every part of an agent separated from the others, so a management team can follow the conversation without writing code. AI 12 oct 2025 Books
Agentic Coding with Claude Code: the value is in chapter one A manual for one specific tool, with the shelf life that implies. The idea that lasts is in the first chapter, and it is not about the tool. AI Craft 9 oct 2025 Books
AI-Powered Business Intelligence: written before the earthquake, still useful How to get machine learning inside the dashboards a company already looks at. It is from 2022, before everyone started talking about something else, and its dull part is the part that still holds. AI 30 sept 2025 Books
AI at the Edge: the question is where it runs, not which model Intelligence on small devices: sensors, cameras, microcontrollers. It left me a criterion I use far from that world, whenever someone wants a model where there is neither a stable connection nor anyone watching. AI Architecture 27 sept 2025 Books
AI and Machine Learning Algorithms for Engineering Applications: a collection with no thread Seventeen chapters by different authors with nothing holding them together but the cover. I include it because weak books teach something too, mainly how to spot one quickly. AI 23 sept 2025 Books
ISO/IEC 42001: who answers when the model gets it wrong A guide to certifying an AI management system. Certification is several sizes too big for almost every company I work with; the vocabulary of accountability is not. AI Strategy 20 sept 2025 Books
AI Engineering: the book that fixes the order in which you spend money Building on models that already exist is a different trade from training them. This book separates the two properly, and its evaluation half is what to read before promising anything. AI 17 sept 2025 Books
50 ML Projects To Understand LLMs: opening the model instead of trusting it Fifty Python exercises that treat what happens inside a language model as data: tokens, vectors, attention, activations. It is not reading, it is work, and it changes what you are willing to promise. AI Craft 11 sept 2025 Books