products and tools I've built

Work

I build AI products for real operational work, along with tools for developing and running them. Explore a workflow, see the decisions behind it, and find out what each project does today.

selected work

Built for real operations

4 projects

Prism portfolio simulation at event six of six, with the result 3 words and an inspect-record excerpt showing two provider turns, one tool call, and completed terminal status.
Portfolio simulation screenshot derived from Prism v0.1.0 public source: run.completed, the result 3 words, and a v1 inspect-record excerpt with fixed limits. No Prism execution was captured; measured usage and cleanup receipts are absent from this CLI record.

Developer preview

Prism

A developer preview for bounded local agent execution, with an inspectable event trace, a contract-derived terminal run record, and fixed limits clearly separated from measured usage.

I build the trace workflow and the interface for inspecting its result.

Explore PrismSource at pinned commitDeterministic getting startedAttribution / NOTICE
Agent Team portfolio simulation at Delivery, showing fictional sample CI and automatic merge, with the default configuration expanded and enforcement settings off.
Portfolio simulation screenshot: a fictional successful run reaches Delivery through the represented default legacy automatic merge path. Merge and queue enforcement, sandboxing, and anomaly termination are off; hard caps are unset. The website performs no repository operation.

Implemented

Agent Team

A software agent workflow for planning, implementation, and review. Its example separates a successful run, a retry, and a blocked result, with the default automatic merge policy stated beside the steps.

I build the orchestration and review workflow around the participating agents.

Explore Agent Team
Agent Config portfolio simulation in a stacked layout, showing a Detailed shared fragment and current Claude Code output followed by the stale Brief Codex output.
Portfolio simulation screenshot with Illustrative outputs: the Detailed shared fragment and Claude Code output remain current while Codex shows the Brief output as Out of date. The stacked layout preserves the complete visible excerpts. No real compiler was run and no source stamp was produced.

Implemented

Agent Config

A shared configuration workflow that turns authored fragments into instructions for different coding tools. The example compares generated outputs and shows what becomes stale when a source fragment changes.

I built and maintain the compiler, runtime, adapters, and integration that turn shared fragments into tool-specific instructions.

Explore Agent Config
Control Center portfolio simulation showing a failed run, a stale run, one unresolved decision, sample spend of $18.40, and a fictional deployment record.
Portfolio simulation screenshot: the Fails scenario and Needs attention filter show two runs and one unresolved decision, with fictional spend and deployment records. Control Center is a Prototype; deployment is unverified. Ages use a frozen sample time. Homepage and Healthchecks retain credit.

Prototype

Control Center

An operations dashboard combining unattended runs, pending decisions, sample spend, and deployment records. This prototype example shows how an operator can find work that needs attention without implying verified deployment.

I built the aggregator, adapters, run wrapper, configuration, and visual composition, with Homepage and Healthchecks credited for their contributions.

Explore Control Center

how i build

How my own operation runs on AI, the tools I actually use, and where I still draw the line at a human.

How my own operation runs on AI

open source

Open source

Tooling I publish on GitHub.

Explore open source

recent notes

All notes
Process

Map it, gate it, check it

My method for putting AI into a business finally has a name: Charted. Map the work, gate the risky parts, and keep checking the output after it ships.

client work

Everything above is mine. Client work is a separate part of this practice. I publish a named client case study only after the engagement is complete and the client has approved what the page says.

See how engagements are structured and what the founding-client offer includes

Have a workflow like this?

Start with the audit

Audits start at $750, fixed scope. Every engagement begins here. Two to three weeks inside your business, and you end up with a map, a scored shortlist, and one recommended pilot, whether or not you hire me to build it.

Discuss a project

Not ready to talk? Take the free AI readiness assessment