First-principles notes on agent systems and AI for traditional businesses.

Writing

Dated, first-person reports on the agent systems I run and build, with the receipts inline. New posts land here without a redesign.

SEP 2026Trellis 1.0: keeping the process when the coding agent changesThe stable release made Claude Code, Codex, and Pi share an explicit process contract. The rollout then showed me where installation, native behavior, and my own confidence still diverged.8 MIN READ JUL 2026GPTx: putting GPT agents inside Claude CodeClaude Code is where I want the work to run. GPTx keeps Claude in charge while selected GPT workers join through the same agent interface, inside the same project session.10 MIN READ JUL 2026Loops need brakes: five Trellis release candidates in five weeksEveryone spent this summer telling you to stop prompting and start designing loops. I agree, and I got there backwards. The brakes came first, then the map, then the gate, then a second engine.9 MIN READ JUN 2026Trellis 1.0-rc: making the process enforce itselfThe last gap was not a missing rule. It was no guarantee the rules ran the same way in Claude Code and Codex, at every layer, on every project. Here is how I closed it.9 MIN READ MAY 2026Renaming six projects on the same dayA retro on the 2026-04-24 mass rename across directories, GitHub repos, hook configs, and scheduled-audit paths, plus the registry-first pattern that made the rest tractable.4 MIN READ MAY 2026Native git hooks for projects without a Node runtimeWhen husky is the wrong tool. Wiring pre-commit, commit-msg, and pre-push enforcement into a Unity game and a Go-first monorepo without inventing a package.json just to install lint-staged.3 MIN READ MAY 2026Stack profiles: where the engineering process bends without breakingHow Lume (Unity) and ClusterBid (polyglot Go+TS+Python) each forced a new stack-profile carve-out, and what that taught me about the Rule of Three.7 MIN READ MAY 2026Folding Anthropic's large-codebase playbook into TrellisA point-by-point gap analysis against the latest Claude Code best-practices guide, and the ten changes I shipped to close the gaps.11 MIN READ MAY 2026Trellis: An Engineering Process for AI Coding AgentsHow I run six projects with one rulebook, nine hooks, and a fleet of weekly audits, and why I open-sourced it.8 MIN READ APR 2026Building this portfolioWhy the site ships with the same process I'd bring to a client engagement. PRD, ADRs, ROADMAP, process-gate, agent-readiness, the whole thing in the open.4 MIN READ APR 2026What I learned building micrograd and makemore from scratchA foundations-first reading of Karpathy's Zero to Hero, and why re-implementing the thing is the only way to understand the thing.1 MIN READ MAR 2026Notes on bringing AI to an MSMEWhatsApp, paper ledgers, Tally, and a few spreadsheets. What actually moves the needle, and what doesn't.2 MIN READ FEB 2026Migrating from Fastembed ONNX to Hugging Face TEIThe specific trade-offs, the retrieval quality delta on an institutional corpus, and why the infra complexity was worth it.1 MIN READ