Continuous Deployment Workflows
summary
Argues that shipping dozens of tiny releases per day is practical by removing release branches and deploy on merge so every merged branch automatically goes to prod, backed by an automated CD pipeline, fast reversible deploys, and comprehensive tests (unit, E2E, visual with Percy, Lighthouse performance checks). Recommends small task branches that merge within 2–3 days, feature flags, and per-branch preview environments using containerized forks of production data (Netlify and DB fork services cited) to get fast user feedback and lower deployment risk.
AI-written summary · about
chapters (10)
- 0:00Introduction and welcome
- 1:47Classic deployment workflow explained
- 4:15Problems with classic workflow
- 10:43What is continuous deployment
- 14:19Splitting work into small steps
- 20:14Requirements for continuous deployment
- 24:22Automated testing in CD
- 35:00Adopting continuous deployment
- 40:38Benefits and closing remarks
- 41:40Q&A session
related talks
44mI Deploy On Fridays
40mContinuous Delivery Anti-Patterns
36mContinuous Deployment with Rails
51mBetter software, faster: principles of Continuous Delivery and DevOps
31mFireside Chat: Continuous Deployment
47mContinuous Delivery Warstories
24mContinuous Deployment at Scale
01h 01mI deploy on Fridays (and maybe you should too)
