Programming language evolution in the AI era
summary
Concrete prescriptions: frameworks generate agents.md file generated by frameworks with project rules and precommit tasks (Phoenix ships one after generating 100 sample apps), MCPs add capabilities like Sentry or GitHub but introduce the Simon Willis little trifecta security risk (private data + external comms + untrusted input), fewer MCP clients and better runtimes win, and integrate agents into the product runtime. Demonstration shows Tidewave as an in‑browser coding runtime that maps DOM to templates, implements a chat search feature, runs browser tests and DB checks, finds and fixes its own bug, and therefore actually works.
AI-written summary · about
chapters (7)
- 0:00Introduction to coding agents
- 2:25Defining agent components
- 3:22Improving agent instructions
- 9:10Agent tools and security
- 15:34Optimizing agent runtimes
- 23:25Integration strategy
- 27:22Introspection and future
related talks
38mCoding Agents & Language Evolution: Navigating Uncharted Waters
01h 03mAI Coding Agents and how to code them
01h 28mAI Coding Agents and how to code them
48mAI Coding Agents and How to Code Them
30mBuilding Tools with Tools: Modern AI-Driven Development
49mFrom autocomplete to agents: AI coding assistance state of play
40mAI Coding Agents and how to code them
27mEmerging Patterns Agentic AI