The Data Praxis

About

"The magic is in the things you are avoiding."

I write these articles to understand a topic better and identify my own gaps in understanding. Publishing forces a rigor that private notes do not: if I cannot explain something clearly enough for another practitioner to act on it, I have not understood it well enough myself. If something here helps you along the way, that is exactly the point.

I'm Vikas Pratap Singh, a problem solver who works in the data and AI space. I combine product thinking, technical architecture, and governance sensibility to solve problems for enterprises. Over two decades, I have designed Data Governance operating models for regulated industries, built Data Quality and observability systems for platforms processing billions of events, and shipped AI-native products where the AI is the product, not a feature. The hardest problems in all of these were never technical. They were organizational, political, and cultural.

Currently, I serve as an Associate Director and Principal Data Architect at Capgemini, leading data and AI initiatives for Fortune 500 financial services clients. Before that, I was a Senior Data Product Manager at Expedia Group, where I owned Data Quality, Data Observability, and the Metadata Management platform. That role shaped a conviction I carry into every engagement: data products are not internal tooling. They are systems that either earn trust or erode it, one pipeline at a time.

I completed the Kellogg Executive Education, Chief Product Officer program, which gave me a language for the gap between technical capability and business outcomes: how to evaluate products, build go-to-market motions, and make the case for investments that do not have obvious short-term ROI. That gap is where most Data Governance and AI Governance programs die.

I am passionate about data privacy and data protection, and I believe in a privacy-first approach to building products. In my spare time, I build tools with Claude Code for my own day-to-day workflows. Building with AI is the fastest way I have found to learn: it forces you to think in examples, not abstractions.

Why "The Data Praxis"

Praxis means the practical application of theory. The data and AI world has no shortage of frameworks, manifestos, and maturity models. What it lacks is honest writing about what actually works when you try to implement them: where governance programs stall, why AI governance is not the same as AI ethics, and what Data Quality means when you are the one debugging the pipeline at 2 AM. This site exists to close that gap.

I am building this as a body of reference work for data practitioners. Not hot takes or trend commentary, but durable writing on best practices in Data Governance, AI Governance, AI Safety, and the engineering decisions behind them. The kind of resource I wish existed when I was figuring these things out for the first time.

What I Write About

The coverage spans eight areas, chosen because they are the problems I have spent my career solving and the ones I keep returning to:

  • Data Governance & Management: frameworks, operating models, and practical implementation
  • AI Governance & Safety: policy, regulation, and safe AI deployment
  • AI Products & Strategy: AI-native products, evaluation frameworks, guardrails, and go-to-market
  • Data Architecture & Engineering: platform design, pipelines, and infrastructure
  • Metadata & Data Quality: cataloging, lineage, observability, and quality at scale
  • Model Risk & MLOps: model validation, monitoring, and deployment
  • Industry Teardowns: deep dives into how leading companies solve data challenges
  • AI Paper Breakdowns: accessible summaries of important research papers

Perspectives

I have been mentoring and coaching people throughout my career, and the conversations that stick are rarely about technology. They are about how to think clearly under pressure, how to build a career that compounds, how to make decisions when the data runs out. Perspectives is where I write about those things: mental models, decision-making, parenting, productivity, and building a career worth having. Honestly, I enjoy writing these more than the technical articles. Keeping both under one roof is deliberate: the way you think shapes the systems you build, and the systems you build reveal how you think.

Connect

I am always interested in conversations about Data Governance at scale, AI product design, privacy engineering, or the messy reality of implementing what the frameworks promise. If something I wrote resonated or got it wrong, I want to hear about it.

Find me on LinkedIn, X (Twitter), GitHub, or reach out at vikaspsingh@outlook.com.

You can also subscribe to the newsletter at the bottom of any article to get new posts in your inbox.

New to the blog? Start here. Curious about the writing process? Read How I Write. For the tech stack, see How This Site Is Built.