Understanding is the new bottleneck

kept by eddie jul 03

As AI agents write code faster than humans can review, understanding remains critical for creative participation, not just verification.

You can just choose how many bugs you want now
AI coding makes bug discovery cheap, letting teams decide how many bugs to tolerate, while fixing them remains costly and complex.
also on Geoffrey Litt, #ai-agents, #ai-coding, #software-engineering, #developer-tools
We Should Take Text Optimization More Seriously
Text optimization—modifying prompts, context, memory, and harnesses—is a legitimate, sample-efficient learning mechanism that deserves the same rigorous study as weight optimization.
also on Notion, Cursor, #ai-coding, #software-engineering, #developer-tools
A Mental Model for Agentic Work
Agentic work follows a universal five-component architecture (LLM, host, loop, context, shared workspace) seen across OpenClaw, Cursor, and Notion.
also on Notion, Cursor, #ai-agents, #developer-tools