louisfb01/start-ai-engineering

This repository is a free, community-maintained guide for anyone who wants to learn AI engineering in 2026, starting from zero programming or AI experience. The maintainer, Louis-Francois Bouchard, runs the What's AI YouTube channel and has organized resources by how people prefer to learn: videos, articles, books, online courses, and hands-on projects. Most resources linked are free, with paid options clearly labeled. The guide covers a wide range of topics that make up modern AI engineering. Those include working with large language models, building retrieval-augmented generation systems that combine AI with external data, designing workflows and multi-agent systems, writing evaluations to measure whether a system actually works, deploying models, handling security and safety concerns, and understanding reasoning models that do extra computation before answering. Each resource carries a difficulty marker from beginner to senior level so you can sequence your learning. The guide comes with a specific point of view: AI engineering in 2026 is not just about writing prompts. It is about the judgment behind the work, including deciding what architecture fits a problem, knowing where a system will break, and determining whether something is reliable enough to ship. The guide suggests using coding assistants such as Cursor or Claude Code to speed up work, while still building foundational understanding rather than delegating all thinking to those tools. A suggested default path starts with short videos to build vocabulary, then moves through a free course, a framework's documentation, a book or two for deeper grounding, and finally two or three small projects that intentionally break. The guide also includes job-finding advice, community links, newsletters, and people to follow to stay current as the field changes quickly. The full README is longer than what was shown.