Hi, I'm Prasanth

Most architecture documentation is three diagrams trying to explain something nobody understood well enough to draw once. I spend a lot of my time now trying not to add to that pile, on Fabric platforms, RAG pipelines, and the standards a team still follows after I've moved on to the next problem.

What I do now

I lead the Data & AI practice: architecture reviews across teams, the standards that get applied project to project, and the client-facing conversations that used to sit above my old title. Increasingly, the job is deciding what “good” looks like before someone else builds it wrong at scale.

How I think

• Provenance is part of the claim. A number without a lineage is a guess with better formatting. I want to know where it came from before I trust what it says.
• If I do it twice, it becomes a tool. The second time I hand-write the same Fabric audit query is the last time. That instinct is roughly how Fabric Lens and fabric-ai-meta both started.
• An architecture that needs three diagrams isn't understood yet. If I can't draw it once and have it make sense, I don't understand it well enough to hand it to someone else.

This blog

I write here to think in public. The articles skew technical: architecture deep-dives, patterns I've found useful, and occasionally the things that didn't work and why. If you build on Azure, work with AI systems, or are just interested in distributed systems at scale, you'll probably find something useful here.

The articles page has everything, or you can browse by topic. There's also an RSS feed if that's your thing.

What I think about AI in analytics

Every semantic model ends up serving three consumers with three different demands: reporting, self-service, and now AI agents. Pretending one model can satisfy all three is where most Fabric implementations quietly go wrong. I wrote the long version if you want the trilemma spelled out.

Building

Get in touch

The best way to reach me is on GitHub or LinkedIn. If you spot an error in something I've written, open an issue or leave a comment. I'd rather be corrected than confidently wrong.

Outside architecture reviews: stargazing, fostering rescue animals, keeping the local pollinators fed, and reading more Greek philosophy than is strictly useful in the job.