Build Log

The proof behind the claims.

This page is the receipts: real builds, described with the same specificity I'd use in a technical scope document. That's the standard I hold the work to, whether or not anyone's grading it.

Operational Vault at AutoBoost

What it is

At AutoBoost I lead client success and growth for 400+ auto repair clients. Before every client review, an account manager used to spend twenty minutes reconstructing context scattered across HubSpot, Slack, and memory. I built an operational vault to fix that: the same second-brain pattern from my personal wiki, scaled for an agency. Markdown vault, CLAUDE.md operating contract, agents that read before they act.

Why it's on this page

This is the proof for the other half of the claim on the home page: the systems instinct runs in both directions, not just toward personal projects. Someone who only builds what's commissioned is a different kind of proof than someone who also runs the operating side of a real business.

Read the write-up

Long Game Studios

What it is

Long Game Studios is where I build things nobody commissioned, on my own timeline, to keep the tools sharp. DailyChew is the current build, in public with Sami, alongside two day jobs and a wedding year.

Why it's on this page

The studio is my evidence of range and shipping cadence: the umbrella for independent products, experiments, and public writing outside of client work.

A Note on Why This Page Exists

Anyone can list "multi-agent architecture" as a skill. Fewer people can describe the failure modes of a LangGraph pipeline at 11pm, or explain why a temporal knowledge graph beats a flat chat history for personalization. This page exists so you don't have to take the claim on faith.