Technologist Describes Walk-Driven Development Workflow Using AI Agents
Geoffrey Litt says long voice notes taken on walks give AI agents more context than a one-off prompt.
363993636537K
“A workflow I'm enjoying: "Walk-driven development" > go on a nice walk outside 🚶 > record a long audio note: ideas, goals, things to build 🎙️ > agent auto-creates docs/tasks, and kicks off cloud coding agents for me 🤖”
Geoffrey Litt@geoffreylittTECH#1740Story Brief
In a post on X, Geoffrey Litt described a personal workflow he calls “walk-driven development”: take a walk, record a long audio note with ideas and goals, then let an agent turn that into docs, tasks and cloud coding-agent work. In follow-up replies on X, Litt said the point is to capture “broad and deep context,” not just a single prompt, and said his setup uses Notion as an all-in-one tool for notes and task creation.
“this is the pattern done right — you didn't remove yourself from the loop, you moved UP it. the human part becomes the thinking + directing (on a walk!); the agent takes execution. that's not 'less human,' it's the human at the altitude that matters. love this.”
Phibrowser
