The emerging skillset of wielding coding agents
summary
Claims that agents require new workflows not old prompts and that Amp can autonomously edit real production codebases and run tests is demonstrated with a live edit to Sourcegraph AMP using its VS Code extension and CLI plus Linear, Playwright and Storybook for fast feedback. Power users spend thousands of dollars a month on inference but gain bigger productivity; Amp spawns sub-agents for search, writes its own tests, uses multi-threaded threads and prefers long prompts and diffs for review, and the demo change succeeds after iterative runs.
AI-written summary · about
chapters (8)
- 0:00Introduction to coding agents
- 4:15Evolution of AI coding tools
- 6:22Agentic AI paradigm shift
- 7:28Design principles for coding agents
- 13:04Live demo of AMP agent
- 21:11Power user patterns and best practices
- 27:52Agents as learning and review tools
- 33:11Future of agent fleets and takeaways
related talks
19mAmp Code: Next Generation AI Coding
30mHow AI Agents Will Change the Future of Programming
06h 22mUsing AI to Enhance Developer Productivity
25mFrom AI Chaos to Production Ready: Using AI Agents to Improve Your Code Quality
25mBeyond the IDE: Toward Multi-Agent Orchestration
26mAI Agents as Your Co-Developer
30mBeyond the Hype: Using AI Coding Agents Today
29mPairing with the machine: How AI Agents are rewriting the rules of software development