Code reviews, for the culture
summary
Company uses SalesLoft workflow where code reviews are the engineering workflow and shape culture: pull requests with labels and emojis, code owners github feature forces architecture approvals for DB changes, pairing and asynchronous reviews run constantly, teams run with three software engineers per product team, reviews catch bugs earlier (chart shows fix time rising dramatically later), promote knowledge transfer and mentorship, PR comments praise instance doubles and note refactor debt with separate tickets.
AI-written summary · about
chapters (4)
- 0:00Start
- 3:30There are four different types of values: • Permission to play values • Core Values • Aspirational Values • Accidental Values
- 24:04Things to Try Think, Does this code accomplish the author's purpose? Ask questions Check it changes follow standard patterns Think about your reading experience Check for coding guidelines and style Check for technical debt. Respond to every comment
- 25:34Things to Avoid Submitting huge pull requests without context Overwhelming with an avalanche of comments Stating opinion as fact. Asking judgemental questions Ignoring toxic behaviors from high performers
