About
I build production AI systems in regulated environments—where getting it wrong is expensive and compliance is non-negotiable.
I've spent 15 years at the intersection of data, AI, and regulation—solving problems where both innovation and compliance are mandatory, not optional.
My career started at Microsoft in 2007, building telemetry systems on the earliest versions of Azure. I learned that data at scale requires both technical rigor and product thinking—nobody cares about your pipeline if it doesn't answer business questions.
That lesson stuck with me through roles at IBM (building clinical AI for Pfizer), SoundCommerce (Customer Data Platform featured by Google), Agora (launching a $5M ARR data product), and Accenture (AI-powered marketing at $100M+ scale).
Today, I lead AI transformation at the FDA with security clearance, architecting GenAI systems in AWS GovCloud that serve federal regulators. It's the culmination of everything I've learned: production AI that passes audits, governance frameworks that enable innovation, and cross-functional leadership that turns technical capabilities into business outcomes.
What I Believe
Compliance Enables Innovation
Most people see regulatory requirements as barriers. I see them as design constraints that force better architecture. FedRAMP compliance doesn't slow you down if you build governance in from day one.
Data Teams Should Generate Revenue
Too many data organizations are cost centers measured by uptime and query performance. The best ones build products customers pay for—that's how you get a seat at the executive table.
MLOps Isn't Optional
Models in notebooks aren't products. Production AI requires drift monitoring, bias detection, reproducible pipelines, and automated rollback. That's the difference between impressive demos and systems stakeholders trust.
Governance as Code, Not Committees
Governance frameworks fail when they're PDF documents and review meetings. They succeed when they're automated checks in CI/CD, data contracts enforced at ingestion, and observability built into every pipeline.
Technical Depth Matters at the Executive Level
I've seen AI governance fail when led by people who don't understand embedding bias, model drift, or training data leakage. Technical depth isn't about coding every feature—it's about making architecture decisions that protect the business.
Ship Value, Not Perfection
I prototype fast, architect for scale, and ship incrementally. Quick wins build trust, strategic bets build the future. The key is knowing which is which.
My Approach
1. Discovery: Understand the Real Problem
I start by talking to stakeholders—not just technical teams, but Sales, Marketing, Product, Legal, Security. What's blocking revenue? Where are compliance risks? What decisions need better data? The best data products solve business problems, not technical curiosities.
2. Prototype: Build Trust Early
I prototype fast—not production-ready, but enough to show what's possible. At Agora, I prototyped Agora Analytics in 2 weeks before getting buy-in to build it properly. At FDA, I demonstrated RAG capabilities before architecting the full compliance framework.
3. Architect: Design for Scale and Compliance
Once there's alignment, I architect for production: governance built in, MLOps from day one, cost controls, observability, and clear ownership. I've learned that retrofitting compliance is 10x harder than designing for it upfront.
4. Ship: Deliver Value Incrementally
I ship in phases. Quick wins first (prove value, build trust), strategic bets next (longer horizon, higher impact). At SoundCommerce, we had dashboards in production in 30 days while building the full CDP over 12 months.
5. Scale: Build Teams and Processes That Last
I hire for "zero to one" skills first, "one to ten" skills later. I establish sprint cadences, code reviews, data contracts, and on-call practices. The goal is a team that can ship without me—I'm most valuable setting direction, not being in every code review.
What I'm Looking For
I'm exploring Chief Data Officer roles in federal government and regulated industries where my experience shipping compliant, revenue-generating AI systems can drive strategic impact.
Ideal roles:
- Federal Chief Data Officer
- Chief Data Officer (healthcare, finance, life sciences)
- VP/Director AI Engineering with path to CDO
- Head of AI Governance & Compliance
What excites me:
- Building data organizations from zero or scaling existing ones
- Production AI in high-stakes environments (regulation, safety, trust)
- Turning data/AI investments into measurable business outcomes
- Cross-functional leadership with C-suite and Board stakeholders
What I bring:
- Technical architecture depth (I've built these systems, not just managed them)
- Regulatory compliance expertise (FedRAMP, HIPAA, security clearance)
- Proven ability to build revenue-generating data products ($20M+ GMV, $5M+ ARR)
- Team building experience (hired, mentored, and scaled teams 3x)
- Business acumen (I speak CFO, CMO, and CTO fluently)
Let's Connect
If you're building AI systems where compliance and trust are non-negotiable, let's talk.