Hruday
Leads vision, protocol architecture, partnerships, and the long-term standardization roadmap.
The Decentralized Backbone for Automotive AI
“Connecting OEMs, Data, and Intelligence via a low-latency, intent-first protocol.”
Modern vehicles can see. They cannot speak. DriveLink is the missing communication layer — a low-latency standard for vehicles to broadcast intent before they act.
From sensor read to broadcast intent — the full DriveLink pipeline, end to end.
DriveLink reads core telemetry — speed, steering angle, brake pressure, lateral acceleration — directly from the vehicle bus. The on-vehicle module ingests this stream at 100Hz.
The prediction engine generates a probabilistic trajectory cone for the next 0.8 to 1.5 seconds. It learns from local driving patterns and refines online as the vehicle moves.
The vehicle's predicted motion and current intent — lane change, brake, merge — are packed into a small DriveLink message and broadcast to nearby vehicles over the mesh.
Receiving vehicles fuse incoming intent with their own perception and adjust planning before danger appears. The result: smoother merges, no brake-waves, fewer blind-spot collisions.
Every milestone, every validation, every credential — chronological.
The founding team identifies the gap: modern vehicles operate as isolated agents. No shared intent means late reactions and preventable accidents. DriveLink is born.
Concept validated by PedalStart mentors Harsirjan Kour (Core Team) and Sayanee Bhowmik (Ex-VC, mentor). Vision refined; direction set.
High-impact pitch deck built; first investor meeting with Debasis Chakraborty (CEO, Dariaan Consulting). Insights from the CEO of Simple Energy on real-world EV fleet movement patterns.
Partnered with a technical team from NMIT for hardware support. Their expertise moved the first autonomous system from theoretical models to a functional, working implementation.
Competed against teams from across India in the Pitch Arena national finals at IIT Delhi.
Pitch Arena Finals · IIT Delhi
Pitched at Manipal Bangalore. Participated in SMVIT — no trophy but technical advice helped refine the architecture and identify critical improvements.
Secured 4th position at a national hackathon — and were awarded a Patent Grant Option recognizing the innovation and originality of the solution.
Hackathon · April 2026
Three co-founders and the technical core — systems, software, hardware, and the ML brain.
Leads vision, protocol architecture, partnerships, and the long-term standardization roadmap.
Builds the prediction engine, simulation environment, and V2V messaging intelligence.
Handles hardware feasibility, integration, mechanical systems, and real-vehicle interfacing.
Architects the RandomForest decision models — lane change, turning, and V2V negotiation — and the real-time inference that drives every car.
Placeholder bio — confirm Harish’s mentorship focus and guidance to the team.
“Investors, OEMs, fleets, and mobility partners — let's set the standard for cooperative mobility.”