DevOps for Java Developers (or maybe against them?!)
summary
Argues DevOps started as ops pushing more work onto developers and then shows concrete remedies: treat production as primary (feature flags, observability, canary rollouts), make developers on-call, and speed feedback by fixing slow builds. Recommends predictive test selection, remote caching, using all CPU cores locally to get a reported 12x speedup (example: 2 minutes to 10 seconds). Cites scale pressures (175 zettabytes), 90% cite slow builds/tests, 65% lack observability, and promotes build scans and cloud-parallel local builds as practical fixes.
AI-written summary · about
chapters (10)
- 0:00Introduction and DevOps conspiracy
- 3:59The Phoenix Project and DevOps origins
- 10:06DevOps infinity loop and developer role
- 15:06Quality and developer responsibility
- 20:56Testing in production and on-call duties
- 24:23Developer resistance to DevOps
- 26:16Productive DevOps and team collaboration
- 33:07Feedback efficiency and build speed
- 38:25Optimizing developer productivity
- 45:12Developer Productivity Engineering resources
related talks
37mDevOps for Java Developers (Or Maybe Against Them?!)
31mDevOps for Java Developers (Or Maybe Against Them?!)
41mDevOps for Java Developers (or maybe against them?!)
44mDevOps for developers (or maybe against them?!)
01h 01m"Doing Devops" as a politically powerless developer
40mDevOps, 12 Factor, and Open Source
55mDevOps isn't just about developing
37mDevOps, 12 Factor and Platform Engineering