Headline

Building at the edge of AI x hardware · IIT Madras

About

I started early in AI - my final-year thesis at IIT Madras led to Predible Health in 2016, where we built Deep Learning systems to help clinicians interpret medical scans faster. We didn't just write algorithms; we deployed GPU rigs on-premise inside hospitals to run them. That 0→1 journey ended in an acquisition by nference in 2021. Since then, I've built software products across different scales and domains. At Bounce, I worked on supply-shaping algorithms that helped commuters reliably find rides - scaling from 10K to 100K+ daily rides in under 3 quarters. At Plum Benefits, I owned the Admin Experience - scaling the core platform for HR teams as we moved from SMBs to enterprise, contributing to a 15× revenue increase. Over time, I started to notice a pattern: we’re over-optimizing software while under-exploring intelligence in the physical world. We’ve perfected dashboards and workflows. But the next leap isn’t another layer of SaaS - it’s systems that sense, decide, and act. I'm now on a deliberate sabbatical, going deep on applied deep tech - exploring how AI can be embedded into machines, not as a feature, but as the intelligence layer that makes them think, adapt, and improve over time. This could look like robotics, energy systems, industrial intelligence, or entirely new form factors. If you're a researcher, engineer, or operator thinking along similar lines - let's talk. When I'm not building products, you'll find me lifting weights, backpacking, or floating weightlessly underwater as a certified diver - just me and my bubbles.

Location

Bengaluru

AI stack

anthropic apis

claude code

claude

cursor

lovable

Work experience

Founder at Stealth Startup, Dec 2025 - Present · 8 mo, Bengaluru, Applied AI for smarter machines and operators.

Associate Director - Product at Plum, Aug 2021 - Dec 2024 · 3 y 4 mo, Bengaluru, Karnataka, India

Senior Product Manager at Bounce, Oct 2019 - May 2021 · 1 y 7 mo, Bangalore, Looking after Supply Shaping - ensuring the bikes are available at the right place at the right time via in-house Assignment and Redistribution engine (for demand fulfillment); seated at the cusp of workflow automation and system of intelligence.

Founding Team and Product at Predible, Jul 2016 - Jul 2019 · 3 yrs 1 mo, Bengaluru Area, India, Built an AI-enabled software to aid practitioners/physicians in reading medical images using Deep Learning algorithms and ascertain relevant abnormalities, which got acquired by Nference.

Business Analyst at Credit Suisse, Jun 2015 - May 2016 · 1 yr, Citypark, Central Business Avenue, Hiranandani Business Park, Powai, Mumbai., Part of an 8-member analyst team, providing data driven and customised solutions for North Americas high net worth and ultra-high net worth individuals (HNWIs and UHNWIs). − Conceptualised dynamic reports by identifying the essential KPIs to achieve the business objective i.e., servicing RMs to provide data driven, customized solutions for high net worth clients − Built a Transition tracking model – daily tracking of transition of assets worth over 100 billion USD of 15,000+ clients, from Credit Suisse Private Banking Americas division to Wells Fargo − Automated the process of procuring data and data massaging, thus reducing the turnaround time and enabling us with deeper understanding of our performance

Summer Intern at ITC Limited, May 2014 - Jul 2014 · 3 mos, Bengaluru Area, India, Indian Tobacco Division (ITD), Bengaluru - Project entailed improving the End Seal Visual Quality of Cigarette Packaging − Designed a metric to assess visual quality of cigarette packets; the metric got replicated across 6 ITC Indian Tobacco Division factories - all over India − Led field survey to develop the metric from the Consumers’ Point of View in order to drive more sales of ITC cigarettes − Used the metric to implement live changes on shopfloor leading to a ~20% improvement in the packaging quality − Introduced Image Processing to automate the metric, removing subjectivity to improve its repeatability & reproducibility, thereby further enhancing the purpose of the metric − Implemented changes projected reduction of 10,000 hours per annum in man-hours for Quality Assessment

Education

Bachelor of Technology (B. Tech), Chemical Engineering, Indian Institute of Technology, Madras

All India Senior Secondary Certificate Examination, Physical Sciences, dav public school

Public links

github: https://github.com/digvjsmahra