Getting Started with OpenObserve
"AI-first" is easy to say and hard to prove. At OpenObserve we ship two AI features to users - O2 Assistant and the AI SRE - and run our own engineering shop the same way, with DocGen writing our docs and the Council of Agents writing and healing our end-to-end tests. This is the story of moving both out of "a human runs this locally" and into CI, where they now fire on their own: giving agents a browser, an ingestion API, and a real running instance so they can build, seed, click, and verify the tedious work instead of asking a person to grind through it.
Discover how OpenObserve built the "Council of Sub Agents" - eight specialized AI agents powered by Claude Code that automate end-to-end testing. Learn how we reduced feature analysis time from 60 minutes to 5 minutes, eliminated 85% of flaky tests, grew test coverage from 380 to 700+ tests, and caught a production bug before customers reported it. This deep dive reveals the architecture, real-world impact, and lessons learned from building an autonomous QA team that doesn't just automate testing - it amplifies quality.
