Reflections · Open source & AIWhat should we open source in the age of AI?As implementations become easier to generate, task definitions, benchmarks, and acceptance criteria deserve more attention. A programmer’s reflections on development and open-source collaboration.September 28, 2026jo32Read the story ↗
Engineering notes · Agent evaluationHow I use benchmarks to improve an agentI traced repeated tool-argument errors in MiMo to the DeepDeck interface, changed it, and compared errors, time, steps, and cost in a retest.September 23, 2026jo32Read the story ↗
