DevOps for the Startup
summary
Startup adopts infrastructure-as-code with terraform, runs CI in Bitbucket Cloud, and keeps master synonymous with production while delaying automated deploys behind a manual gate. Unit tests hit 90% coverage and build time targets sit under five minutes after trimming a 10-minute pipeline. SOC2 constraints force AWS accounts and encryption choices so some simplicity is sacrificed. Datadog and tracing are intentionally deferred as overengineering; cost alarms only trigger a review once spend approaches low five figures.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 1:20Definitions
- 2:14THE STARTUP PROCESS
- 2:41Evolution of Startups
- 4:04Types of Startups
- 4:31Joining an Early Stage Startup
- 7:36What is important? What is valuable?
- 9:12Constraints
- 10:05What to do? Find the pain points.
- 12:00Fast forward a few months
- 13:34Find the next pain point.
- 15:00Reliability
- 16:15What's next for us
- 17:13Conclusion
- 21:02Build times
related talks
36mDevOps in Outsource… Really!
06mDeploy Like a Startup, Scale Like an Enterprise
15mDevOps for Mere Mortals
28mSupport and Tools to Help You Accelerate Toward DevOps
55mDevOps @Scale (Greek Tragedy in 3 Acts)
31mDevSecOps Ain't That Hard, a Framework
31mDevOps in Infrastructure — When It's Mostly Ops...
48mDevops@Scale