Agentic AI consulting · 25+ years building software
I build things.
For over 25 years, I've been building software, launching companies, and shipping open source. These days I spend my time on AI agents — running them in production, measuring what they really cost, and learning where they don't belong. From genetic algorithms to agentic AI, I follow the interesting problems.
What I Do
Consolidate scattered AI efforts
Three teams, three experiments, no shared strategy, and nothing in production. I've done this cleanup before: find the duplicated work, kill it, and leave the org with one direction instead of five.
Cut what your AI is costing you
Most teams are burning several times what they need to and can't see it. This is telemetry, context budgeting, and architecture — not prompt tweaking. I've taken a single task down 70% without losing anything.
Keep quality from drifting
Agents still make bad architectural calls, and the slop compounds quietly until it's everywhere. Quality gates, ticket contracts, and regression tests for work that isn't deterministic — so your standards actually hold when output goes up.
Know where AI doesn't belong
Half of what teams route through an LLM should be a script. Getting that boundary right is usually the cheapest win on the table, and nobody selling you AI will say it out loud.
What I've Been Building
FounderHawk
An AI brand team for founders — five specialized agents handling strategy, content, analytics, coaching, and networking across every brand you build.
NCLEX Kingdom
A mobile NCLEX-RN study app for nurses. RPG quests, boss battles, and 4,000+ practice questions — gamified higher education that actually sticks.
Sandstorm Desktop
An AI agent orchestrator that manages multiple agents in parallel through Docker-based isolation. Open source.
Scribble Cards
A turn-based card battle game on Roblox, built with my son. Hidden hands, tactical swaps, and abilities that break the rules of physical card games.
OnoSchool
A homeschool management app designed for families who take learning seriously but want to keep it simple.
Onomojo
Developer happiness at the center of software development. A platform that captures team sentiment to align work with motivation.
From the Blog
The Coming Revolution in How We Build Software: Why Programming Languages as We Know Them May Be Obsolete
Every programming language we've ever built was optimized for human cognitive limits — small functions because we can't hold much in working memory, descriptive names because we forget. LLMs don't have those limits. So why are we still making them write in languages designed for us?
What I've Learned From Working Remotely for Over 20 Years
Two decades of remote work — from the unreliable-Skype era through the COVID boom — taught me that the real hidden cost isn't productivity, it's your network. Here's why face-to-face still beats video, why video still beats voice, and why remote success comes down to relentless intentionality.