Private Builds and Metrics to survive not ideal scenarios
summary
Argues that running private builds (local or dedicated pre-integration runs, even temporarily manual) catches most frontend/backend integration breakages that pipelines miss, turning $10 local fixes into avoided $10,000 production incidents. Proposes three actionable metrics: time to feedback (measure delay between publish and reported issue), avoidable integration issues (count from retros of bugs that private builds would have caught) and time to restore trunk (ticketed hours spent fixing others’ broken builds). Implements private builds and these counts to find where ownership and pipeline gaps occur.
AI-written summary · about
chapters (7)
- 0:00Introduction and motivation
- 3:59Transition challenges in DevOps
- 10:04Private builds as a solution
- 14:37Three key metrics explained
- 26:13Recap and ownership shift
- 27:37Q&A session begins
- 42:07Closing remarks
related talks
49mCase study of a good enough architecture
46mSix small steps towards continuous delivery
01h 02mTools for simplified large scale systems
37mHaunted Projects Survival Guide
25mInternal Platform as a Product
46mRugged Software Engineering
41mLegacy, for lack of a better word, is good
33mKeep your dependencies in check