Agentic Coding with Discipline & Skill: Practical Techniques and Tips
summary
Builds and debugs production workflows using agent tools like Claude Code and an earlier local-file plugin called Kaguya, claiming hundreds of millions of tokens of experience over 31 months and a shift from the “10,000 hour” to a billion token rule. Demonstrates five concrete techniques: agent-led technical research, breaking features into agent-digestible tasks, letting agents run build-test-fix cycles, managing long-context conversations, and collaborative debugging; shows CLI-as-RAG, a GitHub growth analysis (3,300→4,300 in 17 days), a mortgage data pipeline, a Mac voice-transcription app, and isolating a flaky CI failure to a dependency update.
AI-written summary · about
chapters (8)
- 0:00Introduction and background
- 1:57History with agentic coding
- 4:23Small project examples
- 7:37Mortgage analysis project
- 9:24Medium project examples
- 13:19Large project examples
- 19:03Additional lessons and tips
- 20:28Q&A session
related talks
30mBeyond the Hype: Using AI Coding Agents Today
40mHow To Tame Your Agents?
48mAI Coding Agents and How to Code Them
01h 59mEmerging Patterns for Coding with Generative AI
01h 01m10 tips to level up your ai-assisted coding
40mAI Coding Agents and how to code them
01h 03mAI Coding Agents and how to code them
01h 01m10 tips to level up your ai-assisted coding