Reading this with an AI? Start at /for-agents/ (orientation) or /llms.txt (index).

Skip to content
TODD BROWN

Todd Paul
Brown Jr.

LOGIC
  • Claude Code
  • Agent orchestration
  • Test & eval gates
  • Revenue operations
  • Web platforms
Illustration of a brain: one half grey circuitry, the other half bright paint

At the frontier of building with AI agents.

CREATIVITY
  • Game development
  • Interactive fiction
  • Audio tools
  • Brand & design
  • Writing
Builds Writing Receipts
FOR PEOPLE BUILDING WITH AI AGENTS

Field notes from building production software with AI agents: what works, what breaks, and the receipts for both.

Everything here comes from real builds. Build logs with the test counts and the failures left in. Long-form lessons on evals that lied, agents that ignored their rules, and the gates that caught them, written so you can use them on your own work. And a field guide for spotting when a system, human or machine, starts optimizing the wrong thing. New pieces land here first. Who’s writing this

Receipts

Every source »
896
automated tests passing on Content Ops
Source

state/repo-state.md, last full run at commit f861f98: 839 TypeScript + 57 Python, 0 failing. Measured. Pulled 2026-09-25.

12 days
Sadyr, from blueprint to production
Source

git log, Sadyr repo: first commit 2026-06-18, live 2026-06-30, 63 commits on 8 working days (claims ledger C-01). Measured.

92 of 95
planned website pages drafted by Content Ops in two days of batch runs
Source

state/manifest.yaml B-03 / B-08, claims ledger C-08. Measured, 2026-09-08 to 2026-09-09.

81,707
lines of gameplay C++ in a game still in development
Source

Game development-stats report, 2026-09-20 (git + line count). Measured. Written by Claude Code under Todd's direction; the game is not shipped.

64
capability packages in LEGION, the system I run Adroit on
Source

state/repo-state.md, filesystem inventory across five capability folders (claims ledger C-11). Measured. Pulled 2026-09-25.

Builds

live

LEGION

The multi-agent operating system I run Adroit on: agents do real work, inside rules the tooling enforces.

Claude Code · Node.js · TypeScript · Sevalla · Kinsta · WP-CLI
Read more »live

Content Ops

An evidence-first content pipeline: discovery, knowledge base and drafts, with a human approving every change.

Claude Code · Next.js · TypeScript · Python · PostgreSQL · Sevalla
Read more »live

Sadyr

A governed knowledge platform that answers only from approved sources, source attached — and runs the sales workflow on revenue deployments.

Claude Code · TypeScript · Node.js · HubSpot · Slack · Sevalla
Read more »in development

A game in development

An action RPG in Unreal Engine, built with Claude Code. Playable build under playtest. Not shipped.

Claude Code · Unreal Engine 5.8 · C++
Read more »shipped

Cadence

A proofread-by-ear desktop app I built because I'm dyslexic. Fully local. Shipped to an audience of one.

Claude Code · Electron · Python · FastAPI · an open-source speech model
Read more »shipped

Brain Vault

An ADHD accessibility system: a local knowledge base structured so an AI agent can navigate it.

Obsidian · Python · Syncthing · TypeScript · Three.js
Read more »shipped

Earlier work, 2012–2024

Twelve years of marketing and operations work: brand guides, case studies, websites and dashboards.

Spreadsheets · Branding · Case studies · Design and development
Read more »shipped

Audio Apothecary

A generative audio app for focus and sleep, built in six days. Shipped to my own phone, not a store.

Claude Code · Rust · Flutter · Dart
Read more »in development

Milkman

A local-model benchmarking tool that refuses downloads that won't fit. Built in 31.4 hours.

Claude Code · Python · FastAPI · React · llama.cpp
Read more »parked

Interactive Fiction Engine

An on-device interactive fiction engine: a small model on a phone, inside a harness that decides what happens.

Claude Code · TypeScript · node-llama-cpp · SQLite
Read more »parked

Creature Battler

A Godot creature-battler where my session-timer and decision-log habits started. Engine functional, tested, parked on art.

Claude Code · Godot · GDScript · Python
Read more »parked

Job Search Pipeline

An agent-operated job-search pipeline for a family member. Orchestration platform built, piloted, then deleted.

Claude Code · Python
Read more »parked

On-Device Assistant

An on-device AI resident on OpenClaw. Every security probe passed; it still needed babysitting to be worth having.

Claude Code · OpenClaw · Electron · TypeScript
Read more »

Corrigibility

A field guide to diagnosing systems that drift: what a system really optimizes, and whether it can still be corrected. Software, AI models, companies and the people inside them fail in the same shapes.

III — The Instruments · Chapter 9

Markers and Tests

Signs of health are cheap to fake. The markers worth trusting are the ones only real correction machinery can produce, read now, before the outcome.

Read the chapter »
IV — The Applications · Chapter 11

AI Systems

Specification gaming, sycophancy and eval-gaming: the proxy failure in machinery with no self. Two senses of corrigibility, and instruments for builders.

Read the chapter »

All 15 chapters Or browse the writing »

NOW

This month I'm finishing Sadyr's meeting-booking flow, rebuilding this site, and queuing 32 LinkedIn posts for October and November. More »

Follow on LinkedIn