Guided Journeys
Project-based paths for people who build with AI. Pick a goal, build something real, and follow a checkpoint-by-checkpoint route through the guides that explain each step.
How a journey works
A journey is a project with a finished artifact at the end, not a reading list.
Pick a journey
Each one is a project with a finished artifact at the end — a live URL, a login flow, a database with backups — not a reading list.
Work the milestones
Every milestone has something to build, a prompt or command to run, and a checkpoint that tells you it actually worked.
Read only what you need
Guides are linked at the moment a step needs them, so you learn the concept while you are already looking at the code that uses it.
Beginner
Build and Ship Your First Web App With AI
A small web app running at a public URL that a friend can open on their phone, plus the repo it lives in.
16 min read → 1–2 weekendsAdd Accounts and Login to Your App
Working signup and login, sessions that end when they should, and per-user data that another account cannot reach by changing an ID.
15 min read → 1 weekendKeep Your App Online
An app that redeploys on every push, tells you when it goes down, shuts down cleanly, and can be rolled back in one command.
14 min read → 1 dayFind the Money in Your Cloud Bill
A smaller bill you can prove: the biggest line items identified, idle resources removed, the workload right-sized, and a before-and-after figure from two real invoices.
13 min read →Intermediate
Add a Database Without Regret
A schema you can change safely, migrations in version control, indexes built from real queries, and a backup you have actually restored.
16 min read → 2 weekendsDebug and Refactor the Code You Didn't Write
A reproducible bug fix backed by a regression test, a safer refactor of a tangled module, and a codebase you can navigate without asking the assistant what it does.
15 min read → 1–2 weekendsRun a Security Pass Before Real Users
A documented attack-surface inventory, every default AI security failure checked off or fixed, scanning in CI, and rotating credentials issued.
16 min read → 2 weekendsMake Slow Work Reliable
Slow work moved off the request path into a job queue with retries, idempotency, a dead letter queue, and monitoring that notices a stuck worker.
15 min read → 2 weekendsAdd an AI Feature Without Adding an Attack Surface
An LLM feature in production with validated output, scoped credentials, injection defences, cost limits, and audit logging of every model-driven action.
16 min read → 1 weekendServe Users in Asia Without the Round Trip
An app that answers Hong Kong and Singapore users from a nearby edge with a measured p95, not an assumption — and no mainland filing required.
15 min read → 1 weekendGive Your Coding Agent Its Own Sandbox
A coding agent that runs in an isolated workspace, without your credentials, with restricted network access — and a test that proves the boundary holds.
14 min read → 1 dayRun Two Coding Agents Without Losing Work
Two agents working on one repository in parallel, each in its own worktree, with an explicit merge order and a test run on the combined result rather than on each branch.
13 min read →Not sure where to start?
Journeys teach by building. If you would rather understand one concept first, start with the reference.
Start here paths
The five goal-based entry points on the homepage: build, ship, secure, run containers, and understand the machine.
Open the homepage →Working With AI Coding Agents
The pillar guide behind every journey: choose an agent, write a spec, review every diff.
Read the hub →Search the reference
Hit a term in a diff or an error message you don't recognise? Every guide is indexed.
Open search →