Engineering practices Basic
Most of an engineer’s time is not spent on hard algorithms, but on reading, changing and reviewing other people’s code.
This section collects the habits that make that work cheap and safe. Each topic is a cluster made of an Overview page, a few short pages, and a Common mistakes page with bad → good code.
Which clusters a change passes through
Section titled “Which clusters a change passes through”Loading diagram…
The clusters
Section titled “The clusters”| Stage | Cluster | You will be able to answer |
|---|---|---|
| Writing code | Naming | How do you name variables, tables, APIs and events so the whole team reads them the same way? |
| Writing code | Pragmatic clean code | When do you extract a function, and when is duplication better than the wrong abstraction? |
| Writing code | Common coding mistakes | Which async, type, database, security and performance bugs do developers make most? |
| Teamwork | Git for team projects | Rebase or merge? How do you recover commits lost to a reset? |
| Teamwork | Code review | What do you review first, and how do you write comments so PRs merge fast and stay solid? |
| Teamwork | ADRs and technical docs | How do you record decisions so you do not re-argue them six months later? |
| Operations | Config and 12-factor | One artifact running in many environments; does missing config fail right at startup? |
| Operations | Incidents and postmortems | When production breaks, in what order do you act, and how do you learn from it? |
Quick read: each cluster has its own Common mistakes page. If you only have 15 minutes, read that page first, then come back to the explanations when needed.
Related in other sections
Section titled “Related in other sections”- Writing transactions in code: keep data correct when many requests run at once.
- Secure coding for backends: the vulnerabilities that live in the code you write.
- Unit, integration, e2e and Flaky tests: tests you can trust.