Backend Engineering with Node.js & PostgreSQL

Entities, relationships, and normalisation up to the point where it stops helping. Modelling a payments ledger as the worked example.

Reading a Postgres query plan, choosing indexes deliberately, and spotting the accidental sequential scan.

Isolation levels, race conditions in money-handling code, and why 'read, check, write' loses data.

Resources, status codes, pagination, idempotency keys, and error shapes that clients can actually handle.

Password hashing, sessions versus tokens, refresh strategies, and role-based access that does not leak.

Trusting nothing from the client, bounding request cost, and surviving the day someone points a script at you.

Moving slow work off the request path, retries with backoff, and making jobs safe to run twice.

Integrating Flutterwave end to end: initialising a transaction, verifying it, and processing webhooks without double-crediting.

Shipping schema changes with zero downtime, testing your restores, and knowing something broke before the client tells you.