Engineer cuts 100 bots down to 12 that actually work: stop being your agent's loop
blaizedsouza · x · 2026-11-22
SpaceXAI engineer Lauren Tan's 76-minute talk: most people only ask their bots questions, which is why bots never do anything. She went from 0 bots to 100, then back to 12 — and those 12 get more done, running her work, meetings and even her house lights (they turned on automatically when she joined a call, no manual).
The quoted article adds the framework: almost every "autonomous" setup is a human in the loop — it writes, you check. It breaks down 7 levels from babysitting agents to running a fleet, and what Harness, Loops and Graphs each fix and break.
More from coding & agent
- Garmin Health MCP Server lets you query VO2 max, sleep and training data via natural language — modelcontextprotocol · 2026-11-22
- Stripe MCP connector offers read-only access to Stripe data via API key — modelcontextprotocol · 2026-11-22
- Dev ports Inkling 2 Mini to vLLM, hitting ~200 tok/s decode on 2x RTX 3090 vs 50 on llama.cpp — peaster_ · 2026-11-22
- Six OAuth pitfalls from wiring Claude and ChatGPT into a remote MCP server — Dheeth · 2026-11-22
- Open-source Paperwire MCP lets AI agents send postal mail, certified letters, checks and faxes — QuixiAI · 2026-11-22
- After a year of agentic AI coding, verification—not writing code—became the bottleneck — Positive_Picture-1 · 2026-11-22