Guides
Overview · 152 min total

AI & craft

AI-Native Engineering

Six sessions in plain language for AI-native workflow: job listings, the stack, repo rules, verification, and team habits — built around one ordinary task.

Built for: Developers at any level — new to AI tools, daily agent users, or leads setting team norms — who want a repeatable workflow, not buzzword fluency

Welcome

Software job listings often start in familiar territory, React, tests, code review, and then add words that were rare a few years ago: harness, agent, AI-native.

If that combination has ever made you pause, you are in good company. The question underneath is usually the same: I still know how to build software. What are these new words actually asking for?

This guide is six sessions in plain language. No single tool is required. No secret job title to claim. The aim is a steadier way to read listings, work with AI assistants, and check their output, even while the products keep changing.

You will not memorize a glossary and move on. Each session starts with a question for you, walks through one small situation, and ends by asking you to compare what you believed with what the session showed. The thread through all six is the same ordinary task, refactor a payment helper, add tests, open a pull request, so abstract words land on work you already understand.

By the end, you should notice new things in your own week: when a listing is vocabulary, not a closed door; when one long chat is doing too much; when a correction belongs in the repo instead of your clipboard; when green CI is not enough. Those notices are the point.

What this guide is (and is not)

This is not a list of tricks to sound clever in an interview. It is not a race to use the newest model.

It is six sessions built around one ordinary task, refactor a payment helper, add tests, open a pull request: so each idea stays tied to work you already understand.

Who might find this useful

If you…A gentle place to begin
Have never used an AI coding tool beyond autocompleteSession 1, then Session 2’s first-session walkthrough
Use chat or autocomplete, but results feel hit-or-missSession 2, then Session 3
Use agents daily, but repeat the same correctionsSession 3 and Session 4
Lead a team and want shared habitsSession 4 and 5

What helps beforehand: basic comfort with a repository, clone, run tests, open a pull request. You do not need a paid tool or a specific editor.

The six sessions

SessionWhat you will explore
1. The capability shiftWhy listings feel strange, and why autocomplete is the wrong comparison
2. Stack + your first sessionHarness, agent, orchestrator in plain language, plus one gentle first try
3. Teach the repoHow your repository can teach the AI your house rules
4. Verify before mergeHow to check output with a clear head
5. Team patternsShared rules, review habits, and CI floors
6. Check + carry forwardQuiz, full story, quick reference, and what to try next

When you finish Session 6, a quick reference section there collects the decoder tables and checklists in one place for bookmarking.

Each session opens with a short question to answer in your own words, and closes with a reflection on that same question after you have read the session.

Open Session 1 from the sidebar when you are ready. Read at your own pace.