I build AI your users actually trust.
I help founders and CTOs ship agentic AI that holds in production — reliable, legible, and designed for the humans using it.
You^n.tech
Principal AI Architect
Agent platforms + orchestration
Multi-tenant runtime · 14 composable skills
Identity-aware agentic workspace
Multi-agent OS · production-ready
Evaluation & reliability discipline
Simulation benchmarks · published research
How I Build
Four things that show up
in every system I ship.
20 years of production systems — ERP, SaaS, mobile, WebAR, AI. These aren't principles I wrote for a website. They're what I noticed after the fact.
Systems that explain themselves
I build explainable AI — systems where the reasoning is visible, decisions are auditable, and the human in the loop always knows what's happening and why. Black boxes are a design failure.
Human behavior is the architecture
My MA in Transpersonal Psychology isn't a footnote on my resume. It's the lens through which I design every system. How humans actually make decisions is the spec. The technology is the implementation.
0→1 in weeks, not months
A voice RAG agent in 8 weeks. A full-stack AI OS from a blank repo. I design for production constraints from day one — cost-aware, latency-aware, deployment-aware. Demos that can't ship aren't products.
The builder who stays
I own the full stack: agent orchestration, data models, frontend UX, backend, deployment. I don't hand off. That end-to-end ownership is why the systems hold — because the person who designed the architecture built it too.
Case Study Highlights
Selected work, in brief
Production agent platform — WhatsApp to deployed site in one session
Architected NanoClaw + ProductOps: a multi-tenant agent runtime with 14 composable skills. Trigger a complex product workflow from a WhatsApp message — it runs autonomously, builds the site, and deploys. Built for operators, not demos.
Outcome: zero-to-live in one session · Focus: operator reliability
Read case study →Track Record
Results from the work
Not what the system was supposed to do — what it actually did.
$6M → $0M
Revenue growth enabled by platform consolidation
Four legacy CMSs → one stable platform. Publishing and commerce scaled without infrastructure drag. 9-person global team, 12+ concurrent programs.
Director of Engineering · Evolving Wisdom
< 0 weeks
Voice RAG agent, zero to production
Voice-native RAG with polymorphic retrieval, guardrails, and trust-oriented UX. Blank repo to production under a hard pre-fundraise deadline.
Head of AI · HumanOp
0 months
Budget paid back after 10k-SKU migration
10,000+ audio SKUs migrated to cloud, subscription MVP shipped on deadline. Project budget recovered within six months of launch.
Director of Technology · Sounds True
Published Research
“The Flexagon, a Bridge between the Polarized and Dualistic Dimension Reversing Nature of the Cube and the Octahedron”
The Bridges Organization — Mathematics, Art, Architecture, Culture · 2019
Capabilities
What I'm hired to do
Three things I do exceptionally well. Every tool in service of a system that a real person can trust.
Agentic system design
The architecture layer most teams get wrong.
I design agent systems that hold in production — explicit responsibility boundaries, memory that doesn't leak, audit trails that make the reasoning visible. Not clever. Coherent.
Call me for this when
When you're building an agentic product and need the architecture right from the start — not refactored after launch.
Production AI reliability
The discipline that separates a demo from a 3am system.
Eval design, observability, guardrails, RAG quality — built in at the architecture phase, not bolted on after. The question isn't whether it works today. It's whether it still works after 10,000 edge cases.
Call me for this when
When your AI works in demos but fails or isn't trusted in production — and you need someone who can diagnose the real problem.
Full-stack delivery
Architecture to deployment. No handoffs.
I own the full system — agent logic, data layer, frontend UX, backend, CI/CD. The person who designs it builds it and makes sure it ships. Not a team of specialists who don't talk to each other.
Call me for this when
When you need one senior person who can own the system end-to-end.
The background behind the builds
M.A. Transpersonal Counseling Psychology, Naropa University · B.S. Chemistry & Psychology, St. Olaf College · Six-month Theravada monastic residency, Chiang Mai, Thailand
The combination that makes
the AI work differently
How it starts
What working with me looks like
Every engagement starts with the same thing: an honest conversation about what's actually broken and whether I'm the right person to fix it.
Start with a discovery callResponds within 48 hours · Bay Area / Remote
Discovery call
30 min30 minutes. You describe what you're building and where it's breaking. I ask the questions that reveal the actual problem, which is usually different from the presenting problem.
Assessment
Up to one weekFor audits: 3–5 days of system review, failure-mode mapping, and eval design. For roles: a technical conversation and a clear scope agreement. No decks until there's something real to show.
Build or advise
Ongoing or scopedPrincipal role: full architectural ownership from design through deployment. Audit: a prioritized sprint roadmap with expected outcomes. You always know what's happening and why.
How we work together
Two ways to engage
I typically work as a consultant — focused, scoped, deliverable-driven. Occasionally, a company earns full commitment. Here's what each looks like.
Consulting · Project work
I build the thing with you. A defined scope, a clear deliverable, and I stay until it ships. Typical engagements: AI system architecture, audit and sprint roadmap, agent platform build, RAG pipeline design and hardening.
- Defined scope and outcome — not an open-ended retainer
- Full stack if needed: architecture, data, UX, deployment
- Eval and reliability built in, not added at the end
- Typical engagement: 2–12 weeks
Ideal for
Founders and CTOs who have a defined AI problem and need someone to own the build — not manage a team to do it.
Full commitment — for the right company
If your mission genuinely moves me — if the problem is real, the team is serious, and what you're building is worth caring about — I go all in. This isn't a default mode. It's rare. And when it happens, I treat the company as an expression of my care and intelligence.
- Full architectural ownership from design through production reliability
- Principal or Staff AI role — embedded, not advising from a distance
- I stay to make sure it holds, not just until it launches
- The work becomes personal — because that's the only way I work
Ideal for
AI-native companies building something that hasn't been built before, where the mission and the problem are inseparable.
Thinking
The work behind the work.
The Real Revolution Isn't AI — It's Meaning
We built tools for coordination. We forgot to build them for coherence.
I Didn't Build a Productivity Tool. I Built a Fulcrum.
Most platforms assume the problem is effort. I built one that assumes the problem is balance.
Geometry as Error Correction
What if the shape of a quantum processor was itself a form of fault tolerance?
What the Simulator Confirmed
1,000 steps of noise. One topology barely flinched. The other collapsed.
Bay Area / Remote
Available now
Let's talk
I build businesses and products — and bring them to market.
I typically work as a consultant: scoped, deliverable-driven, end-to-end. Occasionally, a company earns full commitment — when the mission moves me and the problem is worth caring about with everything I have.
Either way: the work is an expression of my care and intelligence. That doesn't turn off because the contract says it should.
Use the contact form — I review every inquiry personally and respond within 48 hours.