Privacy taught me what AI leaders still get wrong.

Customer data platform strategy at global scale.

I worked on the data strategy side of MCDP, the customer data platform enabling personalization at scale. That meant turning business requirements into technical specs, thinking through data flows, consent, identity, retention, activation, and governance.

I also managed a team of data scientists and worked across engineering, legal, privacy, security, data, and business teams. The job was translation under pressure: what the business wanted, what privacy allowed, what the data could support, and what engineering could actually build.

Privacy is not paperwork. It is product architecture.

That is the lesson I bring into AI work now. If a company treats privacy as a final legal review, it has already designed the system wrong. Privacy shapes what data you collect, how you structure identity, what the model can access, what gets logged, what gets retained, and what a customer or employee can reasonably trust.

AI makes the old data problems louder.

• Bad data governance becomes bad AI output.
• Unclear consent becomes unclear model access.
• Messy identity becomes unreliable personalization.
• Vendor promises mean nothing unless engineering, legal, privacy, and business can execute together.

That is why my enterprise background matters. I know how ambitious data projects fail inside large organizations, and I know how to design AI work with those constraints in mind from day one.