Beyond the Hype and Hyperbole
summary
Teams report using GitHub Copilot, Codex, Gemini and Cursor (with scoped cursor rules) to generate code, scaffold tests and ship features; one team claims AI lifts test coverage to 60%+ for new modules while another runs three parallel agent branches as a form of babysitting agents to catch drift and deletions. Supercell runs an internal AI lab for enterprise-safe pilots, and several groups say adoption takes months but produces measurable speedups in boilerplate, tests and specs.
AI-written summary · about
chapters (9)
- 0:00AI tooling roundtable intro
- 3:37AI tooling successes in production
- 10:03Test coverage with AI tools
- 16:12Babysitting AI agents
- 21:10Context rot and adoption challenges
- 24:44Onboarding teams to AI tools
- 27:55Overcoming skepticism and adoption
- 31:54Balancing hype and developer empathy
- 32:14Key takeaways and closing
related talks
57mOne Team’s Journey to Using AI Without the Hype
30mBeyond the Hype: Using AI Coding Agents Today
26mAI Agents as Your Co-Developer
27mSoftware engineering with GenAI
32mBeyond the Hype: A Pragmatic Look at How Developers Actually Use AI Tools
26mFrom autocomplete to agents: AI coding assistance state of play
25mBeyond the Hype: Real-World AI Strategies
50mAI Coding Assistance: The State of Play