allsrc.dev

Learning paths that give you a direction, not an encyclopedia

Ordered series for experienced software engineers. Each one starts from the tools you already use, ties every idea to system design you know, and tells you when not to use it.

A learning path is an ordered series of parts. Each part answers one question and builds on the part before it.

Start AI for Software EngineersBrowse all paths
1 path, 22 parts2 h 19 min of reading
The route throughAI for Software Engineers
  1. Start herePart 01 part
  2. Models and how they are calledParts 1.1 to 2.24 parts
  3. Inside the conversationParts 3.1 to 3.22 parts
  4. Extend the AI you already useParts 4.1 to 4.33 parts
  5. Build your own harnessParts 5.1 to 5.33 parts
  6. Knowledge and beyond textParts 6 to 83 parts
  7. Many agents, many failure modesParts 9 to 102 parts
  8. Prove it and secure itParts 11 to 122 parts
  9. In practiceParts 13 to 142 parts

Start here

The path to read first if you are new to this section

AI for Software Engineers

AI for Software Engineers is a learning path that explains how the AI tools you already use actually work, from the model and the context window to harnesses, retrieval, multi-agent systems, evals and security. Each part maps the idea to system design you already know and says when not to use it.

Intermediate22 partsAbout 2 h 19 min of reading
Start with Part 0See the full path

By the end you can

  • Tell a model, its harness, an agent and a product apart, and work out which part changed when behaviour shifts.
  • Follow a request from pressing Enter to the model API and back, including streaming, the system prompt and token usage.
  • Manage the context window: why long chats get worse, what compaction loses, and when to prune, summarise or persist.
  • Extend coding agents with instruction files, skills, tools, MCP servers, subagents and hooks, knowing what each costs in context.

And 2 more

All learning paths

Ordered series you read one part at a time

1 path published

Start here

AI for Software Engineers

AI for Software Engineers is a learning path that explains how the AI tools you already use actually work, from the model and the context window to harnesses, retrieval, multi-agent systems, evals and security. Each part maps the idea to system design you already know and says when not to use it.

Intermediate22 parts2 h 19 min

More paths are on the way

New paths show up here as soon as they are published. Until then, the articles go deep on single topics.

Browse the articles

Questions

Short answers to what people ask before they start.

Who are these learning paths for?

Experienced software engineers. You should be comfortable with APIs, services and basic system design. The paths do not teach programming from scratch.

Do I need AI experience?

No. The paths start from the tools you already use, not from theory or maths. If you have used ChatGPT, Claude or GitHub Copilot, you have enough to start.

How long does a path take?

AI for Software Engineers has 22 parts and about 2 h 19 min of reading. Read one part at a time; whether that takes you hours or days is fine.

Is it free?

Yes. Every part is free to read, with no sign-up.

How up to date is it, and how are facts checked?

Every part shows the date it was last verified. Facts are checked against the provider's own documentation, and details that change often, such as prices, limits and model names, link to the source instead of being copied.

Can I read the parts out of order?

Yes. The order is the suggested route, and each part builds on the one before it, but every part answers its own question. Use the path page to jump to the part you need.

Start with AI for Software Engineers

22 parts, about 2 h 19 min of reading. Begin at Part 0: Why this series, and how to read it.

Start AI for Software Engineers