Coding Agents
Coding Agents Still Forget: Longer Context Is Not Enough
A developer’s four practices for 200K-context coding agents show that context orchestration, not generation, is becoming the key bottleneck.
3 articles tagged with this topic
A developer’s four practices for 200K-context coding agents show that context orchestration, not generation, is becoming the key bottleneck.
GitNexus (45K stars) plugs into Claude Code via MCP, shifting coding agents' bottleneck from writing code to understanding project structure.
Meta ProgramBench tests AI building programs from scratch. Top models failed, cooling 'AI builds software' hype and exposing benchmark score inflation