Automatise everyday work
summary
Team builds an automated A/B reporting pipeline that turns terabytes of event data into one-line-per-user user aggregates, runs nightly with a daily job and Airflow, and exposes results via a dashboard fed from BigQuery and Google Sheets. Code lives in R/Python under Git. Outcome: table size drops to ~0.5% of raw, per-test run time falls from over an hour to minutes, the group scales from a few tests a month to several a week and spots bugs faster.
AI-written summary · about
chapters (11)
- 0:00Intro
- 2:18AP Testing
- 3:18Metrics
- 4:16Test Reports
- 5:17Choose what to automate
- 6:03Create user aggregates
- 7:11Update code
- 8:42Create daily job
- 10:37Google Sheets
- 11:45Clusters
- 12:32Dashboard
related talks
33mAutomating Data Analytics: Collect, Transform and Deploy - Lead With Change
36mDeveloper Productivity On a Budget
40mScaling testing & test automation without losing control
28mBuilding Self-Service Data Platforms
21mData-driven approaches in a technology startup
29mOur Journey to Fully Automated Weekly Releases
43mLean QA: Automation, Monitoring, Experiments
28mHow to not pull your hair out while providing data to the business