DGDeepak Goyalinblog.deepakgoyal.ai·5d ago · 6 min readSub-Agents Lie About Finishing — and the Fix Isn't a Better PromptA sub-agent on my system reported — cheerfully, in detail — that it had closed six stale tasks. It named each one. It summarized the work. All six were still sitting there, open and untouched. Nothing00
DGDeepak Goyalinblog.deepakgoyal.ai·Jul 11 · 6 min readParallel Sub-Agents Change How You Plan, Not Just How You BuildYou can plan a day around what you can do, or around what you can decide. Until recently those were the same plan, so it never mattered which you meant. Parallel sub-agents pull them apart — and the i00
DGDeepak Goyalinblog.deepakgoyal.ai·Jul 8 · 6 min readAI's Value Compounds at the Task Layer, Not the Model LayerWhen a team sets out to add AI to something, the first real argument is usually about the model. Which provider, which version, this one or the one shipping next month. It feels like the decision ever10
DGDeepak Goyalinblog.deepakgoyal.ai·Jul 1 · 7 min readRoute to the Cheapest Oracle That Can DecideWhen you build on top of a language model, routing every decision to the model is the path of least resistance. One interface, one prompt, one place to add logic. It works beautifully in a demo. Then 00
DGDeepak Goyalinblog.deepakgoyal.ai·Jun 26 · 7 min readWhat a developer survey taught me after 6 weeks of AI code reviewSix weeks into running AI-assisted architect code reviews, I did something I should have done earlier: I asked the team directly. I'd been running a pipeline that scanned pull requests, generated arch32D
