Creating a Learning Culture
summary
Describes a concrete program that turns failures into fixes: running a cloud microservices stack at 15,000 cores, 300 microservices reveals repeat failures, so the team enforces an incident flow—incident, immediate fix, debrief 24–72 hours later, and a checkpoint two weeks after—plus Jira labels and company-wide published debriefs. Targets are specific: identify issue in under five minutes and aim to fix in under an hour, prefer feature flags over mass rollbacks, treat config as code, and fix processes not people.
AI-written summary · about
chapters (8)
- 0:00Building a learning culture
- 1:58Goals for deployments and stability
- 3:43Fear in incident analysis
- 6:53Process flaws in incidents
- 8:56Effective debriefs after failures
- 10:24Trust and follow-up challenges
- 12:50Key principles for debriefs
- 13:54Notable learnings from journey
related talks
06mIgnite: Failure and the Third Way
44mDeveloping a Culture of Reliability and Resilience through Chaos
25mWhat I Learned
52mThe Path to Knowledge — How to Build a Continuous Learning Culture
32mTales from the DevOps Transformation Trenches
26mMission Critical Mindset Building a Culture of Resilience in DevOps Teams
32mThis is Not a Technology Problem: Leveraging Culture and Process to Make the Most...
30mLearning from Learnings: Anatomy of Three Incidents