CS student · Developer · Researcher.
I build systems that protect people —
from potholes on roads to pipelines in government.
I'm a computer science student from India who believes the best software is the kind that prevents problems before they become tragedies.
My interests lie at the intersection of civic technology, artificial intelligence, and scalable software systems. I'm especially drawn to building technology that creates measurable real-world impact.
During SIH 2025, my team built SnapSamasya, a full-stack civic safety platform designed to improve how road hazards are reported and resolved. The platform included a citizen portal and municipal dashboard.
Currently working on Project BridgeX — experimenting with scalable monorepo architecture, FastAPI services, and modern desktop application design using React, TypeScript, and Tauri.
Building a strong foundation in machine learning and deep learning with a long-term interest in research-driven AI systems for civic and public-impact applications.
The goal is not to build AI that replaces human effort — but AI that amplifies human reach.
I'm in the foundation-building phase of my research journey — studying core machine learning algorithms, reading papers, and narrowing down a research direction at the intersection of ML for social good and public systems.
I am currently in the active study and topic-scoping phase of my first research project. My work is centered on ML for social good — exploring how supervised learning and deep learning models can be applied to problems in public systems: civic infrastructure, resource allocation, or community services that directly affect everyday people.
Currently studying regression and classification algorithms — linear models, SVMs, decision trees, and ensemble methods.
Building up from perceptrons to multi-layer networks, backpropagation, and modern architectures.
Applying machine learning to problems in public systems — civic infrastructure, healthcare access, education equity.
Actively reading and summarising papers in ML and civic AI to develop research literacy.
Competitive building is where theoretical knowledge meets the brutal constraints of a deadline. Each hackathon has been formative in how I think about rapid prototyping and team dynamics.
India's largest national-level hackathon. Problem statement: build a system to reduce road accident risk from unrepaired potholes.
Built SnapSamasya — a full-stack civic safety platform with two distinct sides. The citizen portal lets anyone photograph a pothole, submit a geotagged complaint, and track its status in real time. The municipal dashboard gives local authorities a live feed of incoming complaints and tools to assign work orders.
Exploring a submission on accessibility technology — designing adaptive interfaces for differently-abled users, with voice commands, screen-reader-first design, and real-time speech APIs.
A working demo of one feature beats a broken demo of ten. Hackathons force you to identify the single most defensible core.
Judges are rarely engineers. The clearest path from model accuracy to citizen impact wins every time.
SnapSamasya only works because both sides — citizens and municipalities — are in the same system. A complaint that goes nowhere is worse than no system at all.
Two-sided civic safety platform — citizen portal to report potholes and track complaints, plus municipal dashboard for work orders and field team management. Powered by YOLOv8 for AI damage detection.
Hand-coded with canvas animations, scroll-reveal, typed text, and responsive layout. Zero frameworks.
Steam-inspired dark-themed launcher. First serious HTML/CSS project — flexbox layouts, polished hover states.
Terminal word game with letters A I P C R H G. Guess GRAPHIC for instant win. Pure Python.
python image.py Mukul — greeted by a T-Rex via cowsay + sys.argv.
Open to research collaborations, internship opportunities, hackathon teams, and thoughtful conversations about technology and its implications.