Patrick Collison's 'liquid software' experiment: freeform user feedback auto-implemented by agents, 150 improvements in days
zeeg · x · 2026-11-01
Patrick Collison is adding an "improvements" feature to his vibecoded apps: a freeform text input wired to an agent that directly implements whatever users submit. The activation energy for polishing rough edges or adding features drops dramatically — one user shipped 150 improvements in a few days. He calls the pattern a new kind of tool, "liquid software," making apps without it feel rigid and anachronistic.
Sentry's zeeg adds that they're experimenting in MCP with agents automatically sending feedback via the same User Feedback API when something doesn't work as expected, closing the loop between agents and users.
More from coding & agent
- Anthropic publishes Skill authoring best practices: keep SKILL.md under 500 lines — bibryam · 2026-11-01
- Dev builds entire Three.js city with no 3D models using Codex and GPT-6 Astra — OpenAIDevs · 2026-11-01
- Compiler-backed coding agent Benzi hits 82.6% on SWE-bench Verified for $9.26 — DonkeyTheKing · 2026-11-01
- Ronie: free open-source AI agent that makes Ray-Ban Meta glasses actually smart — Odd-Sun6296 · 2026-11-01
- GPT-6.1 Terra builds 3-level game in 6.5 min; DeepSeek V4.1 Flash slower but 36% cheaper — latent_lurker8587 · 2026-11-01
- Theo: switching Opus from xhigh to high reasoning actually works better — andersonbcdefg · 2026-11-01