I’m an AI engineer who builds scalable data infrastructure and the intelligent systems that sit on top of it. These days I’m building Piecewise, a platform for running AI agents through structured, cost-controlled workflows, and helping teams ship agents, RAG systems and LLM features as a consultant.
Most of my work lives at the seam between two things that are usually kept apart: large language models and large-scale data. Designing a lakehouse, training an anomaly detector, wiring up a RAG system with multi-agent workflows — they’re the same problem wearing different clothes, which is getting a messy real-world process to behave predictably.
I started somewhere else entirely. Six years as a Junior Telecom Officer at BSNL, keeping twelve rural telephone exchanges and a district’s worth of towers running, taught me more about reliability under constraint than any course since. I moved into data science in 2018 and haven’t looked back.
Some things I care about:
- Building systems that survive contact with production, not just the demo.
- Making complicated things feel simple, in code and in explanation.
- Staying close to the newest tooling without mistaking novelty for progress.
- Sleeping properly. It’s underrated, and I’m told I’m good at it.
Outside the day job I’m a part-time PhD student at IIT Roorkee, which is a slower and more stubborn kind of problem than anything at work.
If you’re building something in this space, or you just want to compare notes, get in touch.
- PhD (part-time) 2025 — PresentMechanical & Industrial EngineeringIIT Roorkee · Uttarakhand, India
- M.Tech 2016 — 2018Solid State Electronics Materials, Department of PhysicsIIT Roorkee · Uttarakhand, India
- B.Tech 2005 — 2009Electronics & Communication EngineeringUIET, CSJM University Kanpur · Uttar Pradesh, India