Dhruv Hegde, Co-Founder and CEO of NADIR

NADIR · Founder

Dhruv Hegde

Co-Founder & CEO · University of Michigan · Canton / Ann Arbor, MI

SWE @ Visa · startup founder · CS + Mechanical Engineering ADAS · sensor fusion · computer vision · 3,500+ LinkedIn followers

About Dhruv

Dhruv Hegde is the Co-Founder and CEO of NADIR, where he leads product, go-to-market, and the technical architecture behind the company’s shadow-mode ADAS sensor health platform. A Computer Science and Mechanical Engineering student at the University of Michigan, Dhruv grew up in the Detroit-area automotive ecosystem (Canton, Michigan) and has spent years at the intersection of computer vision, sensor fusion, simulation, and production software — building systems that have to work under real constraints, not just in slides.

On LinkedIn he describes himself as a researcher and student founder exploring autonomous aerial and underwater systems, with published work across technology and philosophy. He founded AstroGEN–NSSEA, a nonprofit focused on youth research in engineering and astrophysics. Before NADIR, he built bomIQ (B2B procurement SaaS), shipped Hooply (500+ waitlist in 48 hours), and held research and engineering roles at Visa, NASA, CERN, Princeton University, and the University of Michigan Climate and Space program — connecting Kalman filtering, residual scoring, and conformal drift bounds to fleet-scale ADAS problems.

What we’re building at NADIR

At NADIR, Dhruv’s focus is making continuous sensor drift detection infrastructure that fleets, MSOs, and OEM validation teams can adopt in days: ingest existing telematics, score camera, radar, and LiDAR extrinsic health in shadow advisory mode (no ECU writes), forecast time-to-CRITICAL, orchestrate calibration workflows, and export signed evidence for insurance and compliance. NADIR is the API and SDK layer for ADAS reliability — detect extrinsic drift from existing telematics, orchestrate calibration workflows, and export signed evidence bundles.

He believes the near-term future of automotive safety is not another dashboard — it is API-first calibration intelligence that compounds with every vehicle, route, climate, and repair outcome. NADIR is his bet that silent extrinsic drift becomes as monitorable as engine fault codes, and that a team rooted in Detroit OEM culture and U-M engineering talent is positioned to ship it first.

Experience & background

  • NADIR — Co-Founder & CEO Shadow-mode ADAS drift detection API · fleet pilots · HORIZON 3D · Python + TypeScript SDK
  • Visa — Software Engineer Intern SWE intern · Austin, TX (2026–present)
  • University of Michigan Climate and Space — Research Assistant Climate and space sciences research · Ann Arbor
  • Research internships — NASA · CERN · Princeton · Oakland University Aerospace and physics research · published technical work
  • bomIQ · Hooply · AstroGEN–NSSEA — Founder B2B procurement SaaS · basketball analytics · youth STEM nonprofit
  • University of Michigan Computer Science + Mechanical Engineering · automotive safety systems · sensor fusion

How Dhruv thinks about the future

Dhruv writes and speaks publicly on ADAS, AEB mandates, and fleet safety. He cares about systems that survive production: sensor fusion under drift, stream processing at financial scale, and computer vision pipelines that actually ship peer-reviewed results. He sees NADIR becoming the standard shadow advisory layer between telematics OEMs, repair networks, and fleet safety officers — read-only, pilot-ready, and honest about what is proven versus what is still bench validation.

Outside NADIR, he teaches software engineering and computer vision, publishes at the intersection of physics, philosophy, and systems thinking, and builds open-source quant and vision tooling. He is reachable at dhegde@umich.edu, on LinkedIn, and on Instagram.

Explore the NADIR platform — shadow-mode drift detection, signed evidence, and HORIZON fleet forecasting.

NADIR homepage →

We route every submission to team@nadirai.net and respond within one business day.

Ready to send.