About This Site

AI Study (aistudy.dev) is a practical, engineering-focused study guide to artificial intelligence — written for software developers who want to build with AI, not just read about it.

What AI Study is

AI Study is a structured path through modern AI for working software engineers. It covers the whole landscape a developer actually needs — fundamentals, deep learning, large language models, prompt engineering, retrieval-augmented generation, AI agents, vector databases, multimodal AI, infrastructure, MLOps, safety, and the architecture patterns that tie them together — and it treats every one of those as an engineering subject, not an academic one.

The emphasis throughout is on the things that change a decision: how a technology actually works, what it costs, where it breaks, and when to reach for it instead of something simpler.

17in-depth sections

70+deep-dive pages

0PhDs required

Why I built it

Most AI material sits at one of two extremes. Some of it is too shallow — a tour of buzzwords that leaves you able to recognize terms but not to make a judgment call. The rest is too sprawling — excellent, rigorous, and far too long to finish while you have an actual product to ship.

As an engineer trying to build real things with these tools, I wanted something in between.

The bar I set for every page

Dense enough to be genuinely useful · short enough to read in a focused sitting · opinionated enough to help you decide.

Who it’s for

This guide is written for experienced developers and full-stack engineers moving into AI engineering and LLM application work. It assumes you can already ship software and read code.

It does not assume a research background, heavy mathematics, or a PhD — where theory appears, it’s there because it changes something you’ll actually build or debug, and no further. If you’ve ever called an API and wired up a system, you have everything you need to start.

How it’s organized

The curriculum runs from first principles to production — four arcs, each building on the one before it:

Foundations

AI and machine learning fundamentals, deep learning, and how models actually learn.

Applied

LLM engineering, prompt design, RAG, agents, multimodal AI, and system design.

Production

Infrastructure, MLOps, and the safety and security of real AI systems.

Career

Interview preparation and battle-tested, real-world architecture patterns.

Each section ramps from beginner to advanced on its own, so you can enter at your level. And every page follows the same contract: practical first, code and diagrams over dense notation, and trade-offs stated explicitly rather than buried.

How to use it

You can read the guide straight through, or jump in wherever you need to. The home page offers a few learning paths — for those new to ML, for LLM application developers, for platform and MLOps engineers, and for people preparing for interviews — and the sidebar lets you navigate freely. When a term is unfamiliar, the Glossary is meant to be kept open alongside whatever you’re reading.

A note on accuracy — and feedback

AI moves quickly, and parts of any guide to it will age. I’ve focused on the durable ideas — the mental models and trade-offs that outlast any particular model or tool — and I’ll keep the material current as the field changes.

Found something off?

If you spot anything wrong, unclear, or out of date, I’d genuinely like to know — good craft only improves through feedback. You can reach me from the About the Author page.

Thanks for reading. I hope it saves you time.