Do the magic with DevOps, Open Source and All-remote teams (with Python, of course)

on YouTube ↗27m

" summary

GitLab runs a fully all-remote data platform built with Python that actually ships and scales: an enterprise analytics stack using Airflow orchestration, Snowplow → Snowflake ingestion, Fivetran/Stitch for connectors, DBT and Meltano for transformations, plus Monte Carlo for data observability and Permifrost for Snowflake permissions. The team automates tests and zero-copy-clone feature for cheap prod-like environments, routes alerts into Slack, dogfoods GitLab CI to deploy changes, and supports a 60+ country workforce and 30+ million users while keeping costs and time-to-market low.

AI-written summary · about

" chapters (8)

1. 0:00Introduction and background
2. 3:14GitLab company overview
3. 6:00Data team mission and challenges
4. 9:11Dogfooding and DevSecOps mindset
5. 12:19Tech stack and tools
6. 15:51Rising stars in data tools
7. 19:21Data team roles and projects
8. 21:47Q&A session

" related talks

• 51mKEYNOTE: How We Developing Data Services/Products in GitLab.comRadovan Bacovic · Big Data Conference Europe 2022
• 25mHow we create and leverage Data Services in GitLab.comRadovan Bacovic · DSC Croatia 2022
• 31mHow to Implement DataOps usingEmilie Schario · GitLab Commit San Francisco 2020
• 24mOpening Keynote - The Power ofSid Sijbrandji · GitLab Commit San Francisco 2020
• 29mScaling Across the EnterpriseJoost Evertse · GitLab Commit San Francisco 2020
• 31mAI-driven DataOps: Building data products with a taste of the futureRadovan Bacovic · DSC DACH 25
• 21mDriving Cultural Change to Improve Software Development and CI/CDRoland Heusser · GitLab Commit San Francisco 2020
• 18mData Science in Production Has Never Been So EasyAdam Wang · posit::conf(2025)