Salman Khan

Data Architecture & AI Explainability

Data architecture that makes AI work. AI design that people actually trust.

15+ years across TCS, Axis Bank, and LyondellBasell. Doctoral research in AI explainability and decision support. I work with CDOs, CTOs, and risk leaders in BFSI and manufacturing.

15+ years in production

BFSI & Manufacturing

DBA Candidate, Golden Gate University

TCS · Axis Bank · LyondellBasell

What I do

Where I work best

Data Architecture Audit & Roadmap

Zoom out on your data landscape. Identify structural problems, produce architecture diagram and prioritised roadmap.

For: CDOs, Engineering Heads

Data Foundation for AI

End-to-end data model and platform design for organisations whose AI investment is failing because the underlying data is broken.

For: CTOs, Engineering Leads

Performance Optimisation Advisory

Diagnose and fix slow or expensive Snowflake, Databricks, or Oracle environments. Fast engagements, immediately measurable ROI.

For: Engineering Teams, CTOs

AI Decision System & Explainability

For organisations building internal AI decision tools where trust and regulatory compliance are critical. Grounded in doctoral research.

For: Chief Risk Officers, CDOs

"Most practitioners do one of these well. Data engineering. Architecture. Governance. Production AI. I've done all four - in banking, manufacturing, and fintech. That's not breadth. That's the stack."

From the blog

Thinking


Jan 2025
Why Your Data Foundation Is Silently Killing Your AI Project

Most enterprise AI projects fail not because of the model - but because of what sits underneath it. Here is what to look for.


Nov 2024
Explainability Is Not a Feature. It Is an Adoption Strategy.

After three years of doctoral research on AI trust, one thing is clear: if people do not understand why the system decided something, they will override it.


Sep 2024
I Built a Production AI Platform Solo in 3 Months. Here Is What Actually Took the Time.

Not the model. Not the API. The data contracts, the edge cases, and the architecture decisions nobody writes about.


About

The work behind the work

I've spent 15 years building data architectures at scale - across Big 4 consulting, Indian banking, and US manufacturing. My work spans the full data-to-AI value chain: engineering, architecture, governance, and production AI systems. I'm completing a Doctorate in Business Administration with research focused on what makes people actually trust AI decision systems. I work primarily with enterprises in BFSI and manufacturing.

Read more about my background →

If your data foundation is broken, AI won't fix it.

Most engagements start with a 60-minute architecture review. No pitch. Just an honest assessment of where things stand.

Book a call via Topmate