Notes on Effect, AI engineering, agents, and evals — building an open-source coding agent in public.https://xandreed.dev/https://xandreed.dev/posts/tui-solidjs-no-react/https://xandreed.dev/posts/tui-solidjs-no-react/Streaming agent output is a fine-grained reactivity problem. Why efferent's TUI is OpenTUI + Solid signals instead of Ink, and what the cost model looks like.Fri, 12 Jun 2026 00:00:00 GMThttps://xandreed.dev/posts/persistent-context-tree/https://xandreed.dev/posts/persistent-context-tree/Sub-agent context is a tree worth keeping, not scaffolding to throw away. Resume, branch, staleness briefs, and fan-out under one token budget.Tue, 09 Jun 2026 00:00:00 GMThttps://xandreed.dev/posts/colocated-evals/https://xandreed.dev/posts/colocated-evals/Eval distance is an iteration tax. Why efferent keeps its eval suites in the same workspace as the prompts and tools they judge.Fri, 05 Jun 2026 00:00:00 GMThttps://xandreed.dev/posts/bash-approval-rules/https://xandreed.dev/posts/bash-approval-rules/Every approval dialog should leave a rule behind. How efferent makes its permission prompt rate decay instead of compound.Tue, 02 Jun 2026 00:00:00 GMThttps://xandreed.dev/posts/llm-provider-runtime-selection/https://xandreed.dev/posts/llm-provider-runtime-selection/Provider choice is request-scoped state, not architecture. How efferent routes one LanguageModel port across Claude, Gemini, and OpenAI per request.Fri, 29 May 2026 00:00:00 GMT