Thoughts on AI governance, federal data strategy, and building data products that customers pay for.
[Cover Image]
Federal CDOs and enterprise leaders invest millions in AI governance—policies, committees, review boards. Yet AI projects still get stuck, or worse, ship ungoverned. Here's what I learned implementing FedRAMP High compliance for GenAI at FDA.
[Cover Image]
Most data teams are judged by uptime and query performance. But what if your data platform became a product that generated revenue? At Agora, we launched a $5M ARR data product by asking one question: "What if customers could see this too?"
[Cover Image]
Federal CDOs face a unique challenge: balance cutting-edge AI with Evidence Act requirements, Federal Privacy Council coordination, and OMB oversight. Here's how to do both—lessons from shipping production GenAI in AWS GovCloud.
[Cover Image]
I've built data engineering functions from scratch at 3 companies—Agora, SoundCommerce, and through consulting engagements. Here are the mistakes I made and what I'd do differently: when to hire senior vs. junior, build vs. buy decisions, and the first 90 days.
[Cover Image]
I've built Salesforce-to-data-lake integrations at 4 companies. It's harder than it looks—CDC, API limits, data model conflicts, governance. Here's the playbook for canonical data modeling, event-driven architectures, and patterns that actually scale.
[Cover Image]
Shipping AI in healthcare, finance, or government? You need more than a trained model. This is my checklist for production ML: drift monitoring, bias detection, reproducible pipelines, audit trails, and automated rollback. Based on FDA, Pfizer, and enterprise work.
I write about AI governance, federal data strategy, and building data products. No spam, unsubscribe anytime.
How we built a Customer Data Platform that retailers love—featured by Google Cloud.