Developer: My Elaborate AI Workflows Kept Failing — Just Focus on the Outputs
brandon_galang · x · 2026-11-17
Developer brannongalang agrees with thdxr's take that no one evangelizes their AI workflows — only outputs matter, since 99% of the magic comes from the model. He admits that over the past year he repeatedly fell into AI-psychosis-driven "workflow design" rabbit holes, and nearly all ended with him rejecting what he built in favor of simply checking whether he got the outputs he needed. LLM intelligence is too malleable and versatile to over-constrain your usage.
More from coding & agent
- After a compromise, this engineer moved all AI coding agents off his desktop — rseroter · 2026-11-17
- Compound launches IC Deck Bench, first benchmark for IC-ready PowerPoint agents — peterjliu · 2026-11-17
- Confluent's Kafka Agent Skills Praised for Quality, Ship With Built-in Evals — AdiPolak · 2026-11-17
- itr8 puts Claude, GPT, Gemini and Grok in one group chat to collaborate on your work — cantrell · 2026-11-17
- 1000-agent swarm pybot experiment writeup coming within days — lvwerra · 2026-11-17
- Gemini CLI Fix Stops Custom Model Aliases From Silently Losing Generation Settings — sayaklearner-prog · 2026-11-17