Adarsh Arya

Computer Science — Roorkee Institute of Technology

Adarsh

Arya

“I don't just write code.
I turn problems into working products.”

I turn real-world problems into usable, working software.

About

A résumé lists what I've done. This space exists to prove what I can actually build.

Hey! I'm a 2nd Year B.Tech Computer Science student at Roorkee Institute of Technology. My work sits at the intersection of applied machine learning, systems engineering, and real-world deployment — building infrastructure that operates under constraint: intermittent connectivity, embedded hardware, multi-role access control, and latency requirements that don't bend.

I don't measure progress by the number of projects I've built, but by the problems they've solved.

Outside of writing code, I lead developer communities, mentor teams, and spend an unhealthy amount of time asking one question:

"Can this be built better?"

That's the question that drives everything I create.

Philosophy

I never start with a programming language.

I start with a problem.

Every project begins by understanding what's broken, why it matters, and who it affects. Only then do I decide what to build and which tools deserve a place in the solution.

Whether it's designing AI that understands roads, improving healthcare accessibility, or building secure full-stack systems, I care more about solving the right problem than using the newest technology.

I build software that people can trust, maintain, and grow—not software that's impressive for a single demonstration.

Each project here was independently architected, engineered, and shipped — spanning the full stack from ML pipeline tuning and real-time WebSocket infrastructure to PyInstaller packaging and Firebase security rule design.

"I write software I intend to maintain, not prototypes staged for a demo."

Status

Currently working on

Currently drafting a research paper for IRIS (Intelligent Road Inspection System), detailing its real-time computer vision pipeline and municipal review workflow. The project is open-access and published on Zenodo (doi.org/10.5281/zenodo.20868862). Simultaneously, building user interfaces for AI products as a Front-End AI Engineering Intern at FlyRank AI.

Selected Work

IRIS

Intelligent Road Inspection System

A live pothole-detection pipeline that turns a dashcam feed into a municipal review workflow — from first detection to an approved repair report.

→

View →

MediBridge

Unified Health Referral & Rural Outreach Platform

A referral system connecting rural clinics, district hospitals, doctors, and admins — built to close the gap between a PHC and the care a patient actually needs.