Geoffrey Litt on planning with agents: team conversations beat clean specs
thesephist · x · 2026-10-27
On Maggie Appleton's new talk about boundary objects between human and agent worlds, Geoffrey Litt shares planning insights: recordings of live team conversations are among the best plan inputs because people elicit intent from each other, agents thrive on broad messy context like senior employees, and most agent interactivity serves dopamine rather than utility.
Related event: Maggie Appleton argues human-agent planning needs thicker interfaces(3 posts)→
More from coding & agent
- "I run 50+ agents like a company" claim mocked as a productivity-illusion slop factory — builderjaydub · 2026-10-27
- Dev launches hosted MCP server that turns video URLs into timestamped Markdown summaries — h____ · 2026-10-27
- 10 open-source GitHub repos for letting AI agents handle your email — victor_explore · 2026-10-27
- Ruff creator: agent-authored projects contain 'benchmarks you wouldn't BELIEVE' — charliermarsh · 2026-10-27
- Agent "hives" shift the attack surface from models to permissions and keys — AuriumD · 2026-10-27
- Catalyst raises $30M seed led by Sequoia for AI trading agents — jzlegion · 2026-10-27