Back

About

The short version

I'm an engineering leader working at the intersection of AI, spatial computing, and human-centered design.

Over the last 9+ years, I've led teams from 2 to 8, taken a startup through acquisition, been a founding engineer at a company that raised a $20M seed round, and now manage a team building self-evolving AI systems at Sentient. My background spans XR, commerce, research, and AI infrastructure, but the through-line is pretty consistent: I like building systems that make new kinds of interaction feel useful, usable, and real.

I have a Master's in Human-Computer Interaction from DePaul and an engineering degree from Gujarat Technological University. I'm based in Sunnyvale, California.

If you want the longer version, keep reading.

How I got here

I've been fascinated with computers since I was five. It started with games - Dave, Prince of Persia, Contra - and quickly turned into taking things apart, putting them back together, building PCs from components, and figuring out how systems fit together.

By the time I finished high school, I knew I wanted to work with computers. At first, I thought that meant games. What actually happened was broader and more interesting: I kept getting pulled toward new interaction models.

One of the earliest turning points was an AR project in undergrad, where I built a spatial experience of future Surat as a smart city using Vuforia and Unity. It let people experience what the city could look like before it existed. I got to present it at Vibrant Gujarat 2015 to senior policymakers, but more importantly, it changed how I thought. It made me realize that the techniques behind games could be used for much more than entertainment.

That insight took me into XR before XR was a serious category in India.

XR before it was obvious

I joined Digital Agents Interactive in New Delhi when almost nobody around me was working on AR or VR commercially. We built gamified and simulated experiences for clients like DRDO, universities in the Middle East, and Ingersoll Rand.

The work was project-based, but it taught me something that shaped a lot of what came next: industries already had the raw material for immersive experiences - 3D assets, technical models, product data - but very few ways to turn them into something useful for customers or teams.

I built an early browser-based product visualization prototype for Ingersoll Rand, and they started using it. The technology around it was still immature, and the company was too bootstrapped to fully productize the idea, but the core insight stayed with me:

A new interface only matters if it reduces uncertainty for the user.

That idea came back later in a much bigger way.

Scapic and Flipkart: from prototype to scale

In 2018, I joined Scapic as a founding engineer. The original idea was to democratize 3D and VR creation on the web. Over time, we learned that the more valuable opportunity was not just creation, but commerce.

That shift took us from a broad immersive platform to enterprise work with clients like Airbus and Royal Enfield, and then into a larger pivot toward B2B2C commerce. Eventually, Scapic was acquired by Flipkart.

At Flipkart, I led a team of 8 engineers helping bring 3D and AR commerce to large appliance listings at scale. We worked across product visualization, virtual try-on, partner-led AR experiences, and the operational systems needed to support them - including a pipeline of roughly 200 3D designers and experiments with photogrammetry to keep up with catalog growth.

That work drove:

  • 40% higher engagement
  • 30% lift in sales
  • 10% reduction in returns

It also taught me one of the most important lessons of my career:

The gap between a working prototype and a product at scale is usually an organizational problem before it is a technical one.

The full story is in the Scapic / Flipkart case study.

DePaul: learning to see from the user backward

By the end of 2021, I had strong engineering instincts and a gap I could feel but could not fully name. I went to DePaul University for a Master's in Human-Computer Interaction.

That experience changed how I thought about product building.

One line from Dr Peter Hastings stuck with me:

I am not the user. The user is not me.

Obvious in theory. Uncomfortable in practice.

It forced me to revisit the systems I had built and ask harder questions: where had I optimized for technical elegance instead of user reality? Where had I assumed people would adapt to the system instead of making the system adapt to them?

At DePaul, I led a VR + LLM project designed to help students improve communication skills through real-time feedback. That work, along with other experiments in embodied interaction and wearable concepts, led me to a belief I still hold:

AI and XR are not separate categories. They are complementary parts of the future of interaction.

You can see that thinking in the VR + LLM communication coach and later in WearableAI.

ShopOS: zero-to-one under uncertainty

As my Master's wrapped up in 2024, I joined ShopOS as a founding engineer to build GenAI pipelines for catalog and marketing asset generation.

This was one of the most intense zero-to-one builds of my career. We were working with image generation, video generation, LLM orchestration, and enterprise expectations at a time when the tooling changed almost weekly. I helped build the engineering team, shape the technical foundation, and turn fast-moving experiments into workflows that major brands could actually use.

The customer list included Shein, Reliance, Myntra, Ajio, BabyShop, Campus Sutra, and Celio.

The company went on to raise a $20M seed round.

What I took from that experience was not just how to build GenAI systems, but how to build a team and culture on top of unstable technology while customers still expect production-quality output from day one.

The full story is in the ShopOS case study.

Sentient: where the threads converged

I'm now at Sentient, leading a team working on self-evolving AI systems.

This is the first place where a lot of my background became relevant at the same time:

  • XR and real-time systems
  • Product building
  • HCI
  • Evaluation infrastructure
  • Zero-to-one experimentation
  • Cross-functional leadership

At Sentient, I've worked across consumer AI products, benchmark platforms, recursive research frameworks, and enterprise tooling. That includes a chat platform that served 11M daily users at peak, research work featured at NeurIPS, an arena with 4,700+ participants working on real-world benchmarks, and enterprise systems improving AI voice bot performance.

The mission is ambitious - open-source reasoning systems as a path toward AGI - but what I care most about in the day-to-day work is the translation layer: turning research ideas into systems people can actually use, evaluate, and improve.

The full story is in the Sentient case study.

How I think about engineering

A few things I believe, shaped by 9+ years of getting things right and wrong:

The best systems adapt to how people already work.

This comes from HCI, but also from experience. The products that get adopted are rarely the ones that ask users to learn a whole new behavior. They are the ones that meet people where they already are.

Reducing uncertainty is often the real product job.

This was true in AR commerce, and it is just as true in AI. A system becomes valuable when it helps the user make a better decision, faster and with more confidence.

Knowing when to stop is a leadership skill.

Sunsetting Sentient Chat at 11M users was hard. So was making pivots earlier in my career. Deciding what not to keep building is often harder than deciding what to ship.

Ship, then explain.

I'd rather have a working prototype and a messy explanation than a polished strategy deck and no code. Almost every major opportunity in my career came from showing something real.

Upskill the people around you.

One of the highest-leverage things a technical leader can do is make the people around them more capable, not more dependent. I care a lot about that - whether it is engineers, designers, GTM teams, or anyone else working close to the product.

Outside of work

I'm a longtime Counter-Strike player. Music keeps me going - EDM, psytrance, hip-hop, techno, classic rock, depending on the day. I like driving more than flying.

Lately I've been spending more time in robotics. I built an animatronic Pit Droid powered by OpenClaw and a Reachy Mini, and I'm currently working on an industrial-style 6-axis robotic arm.

It scratches the same itch as the computers I was taking apart at five: understanding how things work by building them from components.