How to Optimize Non-Coding Time
summary
Claims saving about an hour a day by automating non-coding work: correlating GitHub to Jira in Slack to do one-click ticket creation, surfacing estimated review time for each PR with a simple ML model, and adding quick merge in Slack and approve snippets. Data shows an average end-to-end cycle of ~6 days, two thirds of that in PR review, half of PRs idle for half their lifespan and a third idle ~80% of the time. Recommendations: keep PRs small (aim <200 LOC) and triage heavy ones into synchronous reviews; Java changes take longer per line.
AI-written summary · about
chapters (4)
- 0:00Start
- 1:13Everyone wants to talk developer productivity!
- 6:14The Pull Request Paradox
- 30:30ML Interlude: Estimating Review Time
related talks
47mHow to Optimize Non-Coding Time
42m7 Metrics to Accelerate Delivery without Damaging Culture
29mBreaking the PR Review Bottleneck
57mAsync Code Reviews Are Choking Your Company’s Throughput
06m50 Haunting Shades of PR Dark Matter
56mAsync Code Reviews Are Choking Your Throughput
14m50 shades of PR dark matter
26mA State of Continuous Merge: The Secret to More Productive Dev Teams