About

I am building at the intersection of robotics, AI, and physical infrastructure.

Over the past 8+ years, I have worked across the full stack of robotics — embedded firmware, computer vision, navigation, mechanical design, simulation, and production reinforcement learning — in three countries. I go deep where it matters most: my work in multi-agent reinforcement learning at Nimble AI was a first in the industry.

At Nimble AI (board: Fei-Fei Li, Sebastian Thrun, Marc Raibert), I built the reinforcement learning division from the ground up as an early employee. The multi-agent systems I deployed improved warehouse throughput by 41% and processed 10,000+ orders daily for Puma and FedEx. These systems spanned fleet coordination, robotic manipulation, and bin packing — five production models running across multiple US warehouses. Collaborating with researchers at Stanford and EPFL, I helped Nimble become a unicorn.

Before Nimble, I built the navigation and perception stack for robot fleets at GreyOrange in India and architected autonomous forklift navigation at Aitech Robotics in Singapore. Earlier, I developed embedded systems across defense, real-time operating systems, and MIT Media Lab prototypes.

I grew up around steel plants and manufacturing facilities in Gujarat — that early exposure to physical systems at scale is what drew me to robotics. From co-founding campus ventures to winning startup competitions, I have always assembled high-performing teams around hard problems.

I hold an MS in Robotic Systems Development from Carnegie Mellon University’s School of Computer Science. At CMU, I published “Classroom Digital Twins with Instrumentation-Free Gaze Tracking” at CHI 2021 and developed meta-learning techniques for robotic manipulation. My research also explored behavioral cloning for autonomous driving simulation.

I studied Electrical & Electronics at BITS Pilani, where I co-founded SmartCampus, a campus-wide IoT platform, and built a humanoid robot with a seed grant from the Director’s funds. I won Best Start-up Idea at ATMOS'13 for an IoT-based energy saving solution, led the sensors subsystem for the AAS/NASA CanSat Competition competing at finals in Texas among ~40 international teams, and studied entrepreneurship through New Venture Creation course.

I prototyped at MIT Media Lab workshops, including anterior segment imaging for eye disease diagnosis (3rd prize, BioAsia Healthcare Devthon) and a wireless mesh network for gesture-based audio. I completed courses in design from the National Institute of Design (Ahmedabad) and robotics from ETH Zurich. I earned Gold Medals in the National Science Olympiad in 2009 and 2010.