Opening
summary
Global DevOps community faces concrete tradeoffs: AI boosts individual productivity (Google quotes ~25–30% generated/accepted code) yet increases instability (DORA notes ~1.5% throughput drop and ~7.2% stability drop in earlier data) and demands data engineering and task decomposition to avoid the “AI doom loop.” A working stack example surfaces: reconciliation loops in Puppet/Kubernetes, then Gemini Code Assist + Gemini CLI using MCP for Jira/Confluence, 1M‑token context and CI that builds, runs unit tests, creates PRs and auto‑reviews — demo shows tests pass and commits pushed.
AI-written summary · about
chapters (4)
- 0:00Opening remarks
- 15:42Keynote on pipelines and AI
- 48:29Keynote on AI reshaping SDLC
- 1:22:41Coffee break and tracks overview
related talks
39mDevOps in the Age of AI: Human Powered Evolution
20mDeveloper Productivity in the Age of GenAI
26mBeyond the code: Leading engineering teams in the GenAI era
38mGenerative DevOps
47mAI, Automation & DevEx: Fueling High-Velocity Engineering
29mAI-Driven Development: Revolutionizing Productivity, Software Development Life Cycle
36mDeveloper Experience Trends & Predictions
19mThe AI Code Review Bottleneck: Escaping the Velocity Trap