AI product systems
LLM evaluation, retrieval-augmented generation, model routing, guardrails, and human-in-the-loop design.
See the document AI caseCapability is never the hard part. Trust is.
My foundation is a BTech in Computer Science and Engineering from Amity University. I started in engineering, building data pipelines and operational systems. That work put me close to the things that make software difficult to rely on: messy inputs, brittle integrations, and failures that stay hidden until someone makes a decision from the wrong data.
Moving into product management gave me a wider view of those problems. At Filo, I helped launch the US tutoring marketplace, where matching logic had to adapt to different supply and student expectations. At Supreme Components, I worked with sales and engineering on an AI-assisted quoting workflow. Knowing what the model could extract was only part of the job; we also had to decide what it should be allowed to send.
I later founded Closphere. I went door to door speaking with business owners and warehouse managers across India. More than 50 conversations changed the company’s direction: customers wanted their existing inventory systems to agree, and had little interest in replacing them. We built an intelligence layer over those systems, reached 63 paying customers, 1K+ users, and exited through a technology/IP sale.
At ProductSquads, I returned to the same question in document AI: how do you make a system reliable across inputs you have not seen? I worked on processing architecture, evaluation, model routing, and human review for a platform handling 100K+ documents each month.
I completed my MS in Information Management at the University of Washington in August 2026. Along the way, I explored product decision support through NXTai and therapist–client matching through Kindred. Building prototypes keeps me close to the decisions that are easy to leave abstract in a roadmap.
As a technical product manager, I want to build AI platforms and enterprise software where reliability, customer adoption, and business outcomes have to work together. I enjoy getting close enough to a problem to understand where the system breaks, and working across disciplines to do something about it.
LLM evaluation, retrieval-augmented generation, model routing, guardrails, and human-in-the-loop design.
See the document AI caseSQL, Python, API and system design, data pipelines, and AWS/Azure. A foundation for working through architecture and delivery trade-offs.
See the inventory platform caseProduct, engineering, and building a business.
Lead Product Manager
Document intelligence, evaluation, model routing, and human review at 100K+ documents/month.
Founder & Product Lead
Built inventory intelligence for 63 paying customers, 1K+ users and exited through a technology/IP sale.
Product Manager
AI-assisted quoting with account context and commercial guardrails across 8K RFQs/month.
Product Manager (Founding US PM)
US marketplace launch, matching, activation, and session experience.
Master of Science in Information Management
Product and Artificial Intelligence
Seattle, WABachelor of Technology
Computer Science and Engineering
Noida, IndiaFor senior PM and AI product opportunities, or a complex problem worth working on.