Welcome to The Data Praxis

The Gap Between the Slide Deck and the System

A few years into my career in data, I started noticing a pattern in every governance meeting I sat in.

Someone would reference Data Mesh. Someone else would bring up data contracts. A third person would mention data products, usually with a slide borrowed from a Thoughtworks presentation or a Zhamak Dehghani talk. The concepts were real. The enthusiasm was genuine. And in almost every case, the conversation stalled the moment someone asked: “So how do we actually implement this?”

The gap was not between smart people and dumb people. It was between theory and practice. Between what sounded right in a conference keynote and what survived contact with a real data platform, real organizational politics, and real engineering constraints. I saw this gap at Expedia, where I worked on data quality and metadata platforms. I see it now at Capgemini, where I work with enterprise clients across industries. The pattern is consistent: practitioners are hungry for analysis grounded in what actually works, not what sounds good in a vendor pitch.

That gap is what pushed me down a rabbit hole. I started paying closer attention to the companies that operate at a scale where governance decisions have measurable consequences: Netflix, Airbnb, Uber, Anthropic, OpenAI. When Netflix published how they govern data across their mesh architecture, I wanted to understand why that approach works at their scale. When Uber published their data quality platform, I wanted to know what problems it actually solved and where it fell short. When Apple claimed privacy was a fundamental human right, I wanted to see where that claim held up and where it broke down.

I was doing this research anyway. The Data Praxis is where that research takes a structured form so other practitioners can use it.

What You Will Find Here

The blog covers three areas, and they connect more than you might expect.

Teardowns of real companies. Not product reviews. Not “five things Netflix does right.” Structural analysis of how specific companies handle Data Governance, privacy, AI, and platform architecture. The goal is always the same: what can a practitioner learn from how this system is built? The Netflix Privacy Policy Teardown is a good example. So is the Uber Data Quality Platform Teardown.

Frameworks you can implement. Governance maturity models, privacy program architecture, AI oversight frameworks, metadata management strategies. These are not theoretical. Every framework on this blog is grounded in specific enforcement actions, regulatory text, or operational patterns from companies that have built these systems. The Privacy Program Framework synthesizes lessons from the Netflix and Apple teardowns into an eight-component architecture you can build from.

Perspectives on thinking and career. Not every important decision in a data career is technical. How you think about systems versus goals, how you apply inversion to hard problems, how you stay relevant when your skills have a half-life: these shape your trajectory as much as your architecture decisions do. The perspective essays are shorter, more personal, and deliberately outside the governance lane.

Start Here: The Data Privacy Series

If you are new to the blog and want to see how these pieces connect, start with the four-part Data Privacy series:

1. Netflix Privacy Policy Teardown: What 325 million subscribers actually agreed to. Scores 38/100 on privacy, 23.7/100 on readability. A case study in the gap between legal disclosure and genuine user understanding.

2. Apple Privacy Teardown: When privacy is the product, where does it break down? Siri’s decade of undisclosed human review, iCloud in China, and a $7.4B advertising business growing behind ATT.

3. How to Build a Privacy Program in the Age of AI: A practical framework synthesizing what Netflix gets wrong and Apple gets right. Eight components, from Data Classification to governance operating models, with AI data as a first-class category.

4. The Data Privacy Regulatory Landscape in 2026: GDPR has crossed EUR 5.65 billion in fines. Twenty US states have privacy laws. The EU AI Act is phasing in. And a new insurance market has emerged for when your AI agent goes off script.

Each article builds on the one before it. Together, they form a progression from “what is broken” to “what is possible” to “how to build it” to “what the law requires.”

What Is Coming

New articles publish regularly. Company teardowns, governance frameworks, regulatory analysis, and perspective essays on career and decision-making. Each article builds on what came before. Series connect across topics: a teardown reveals a pattern, a framework makes that pattern actionable, a regulatory piece shows where the law is heading.

If you build data systems, govern AI models, or make decisions about privacy and architecture at scale, this blog is written for you. Start with whatever interests you. Everything connects.