How we turn data teams into better engineers at Netflix
summary
A small team of eight software engineers enables a 400‑person data org by building tooling and running platforms so data practitioners focus on analytics not infra They create Data Flow a CLI that auto‑creates repos CICD unit tests and branch deploys now managing 300+ repos with ~300 MAU and 50 DAU They host web apps on Posit Connect ($65k/yr) and evolved orchestration from Meson to Maestro following principles make it easy to do the right thing and meet teams where they are
AI-written summary · about
chapters (15)
- 0:00Team and data org intro
- 2:55Enablement team principles
- 4:18Engineering tools limits
- 7:10Data Flow for pipelines
- 9:38Analytics app deployment
- 14:11Staffing and culture
- 16:39Three takeaways
- 18:01Q&A features/problems
- 19:09Q&A career paths
- 20:11Q&A tech stack migration
- 22:16Q&A mindset changes
- 23:19Q&A build monitoring
- 25:03Q&A features vs standards
- 26:09Q&A biggest challenges
- 28:17Closing thoughts
related talks
44mHow Metaflow Became Netflix's Beloved Data Science Framework
03mInterview
54mNetflix’s Journey to Confident Automated Changes
42mBringing Software Engineering Rigour to Data, DataEngBytes 2022
47mData - The Land DevOps Forgot
20mDeveloper Productivity in the Age of GenAI
04mInterview
49mData Platform Architecture, Evolution, and Philosophy