Build a DevOps Culture: Microsoft's Journey to adopt an Agile Mindset and DevOps culture
summary
Microsoft moves 90,000 engineers to deliver updates multiple times per day while the Azure DevOps team ships features every three weeks and grows from 22 features in 2012 to 364 in 2017. Engineering teams adopt aligned autonomy with 41 feature teams owning production, use Git + feature flags and automated Azure Pipelines to eliminate partial automation, and enforce a surprising quality rule: a bug cap = 4 × team size that forces bug sprints.
AI-written summary · about
chapters (26)
- 0:00Intro
- 0:52What is Tupelo, MS world-famous for?
- 1:45Accelerating Delivery with DevOps
- 3:01Azure DevOps
- 6:35The journey so far
- 8:57Five habits we've learned so far
- 9:45Listen to our customers
- 12:53You get what you measure Don't measure what you don't wan
- 15:17Build-Measure-Learn
- 16:10Failure is an opportunity to learn
- 23:54Be Transparent
- 25:59No such thing as 'partial automation
- 27:38Automate completely
- 28:12Security Mindset - Assume Breach
- 32:36Agile at Scale with Aligned Autonomy
- 36:05Teams
- 38:31Planning
- 41:01How do you stay in sync?
- 42:24Sprint Mail – Per Epic (Current Practice)
- 43:27Managing the pipeline
- 44:58Release Flow Using Trunk Based Development to avoid Merge Debt
- 47:17Tests against pull request
- 47:38CI: Green Means Green, Red Means Red
- 48:35Created a "Non-Denominational" Test Taxonomy
- 49:10Test portfolio over time
- 56:16What's next?
related talks
44mDevOps at Microsoft: Our Internal Transformation Story
37mFrom Waterfall to Agile
49mOur DevOps journey – Microsoft’s internal transformation story
01h 01mAgile Evolution: An Enterprise transformation that shows that you can too
07mWhat we learned moving 65,000 Microsofties to DevOps...
01h 02mEffective DevOps for Organizations
01h 01mPragmatic DevOps - How and Why
53mKeynote: Enterprise transformation (and you can too)