From Hero to Zero
summary
Wix moves from chaotic on-call to near‑zero errors by shipping visibility and automation: New Relic and Grafana dashboards, an SLO system with zero error alerts, an alert enricher that prediagnoses incidents, a WildMoose AI agent hitting ~80% root‑cause accuracy, and the Nile framework that generates DB, RPC, validation and testing boilerplate. 4,000 microservices, 600 releases/day and 750 alerts/day drop as 180 services (4%) that cause 98% of errors get cleaned, removing 80% of errors and reaching ~99.995% in three months.
AI-written summary · about
chapters (6)
- 0:00Introduction and journey overview
- 4:34From chaos to visibility in production
- 10:42Automating troubleshooting and incident handling
- 18:35Shifting from thresholds to zero errors
- 25:46Cleaning legacy systems and reducing errors
- 31:34Summary and key takeaways
related talks
25mFrom Zero to AI Hero
32mYour manager’s greatest fear
11mThe Early Agent Gets the Worm
43mFrom Developer to Development Manager
58mFrom AI Sceptic to AI Sceptic Who Ships
32mShipping at the Speed of Life
28mWhat Enterprise IT Can Learn From a Startup’s Journey
39mConfessions of a Systems Engineer: Learning from My 20+ Years of Failure