Hey, I'm Sheel.
I just finished my Computer Science degree at TMU and I'm already thinking about what's next. Mechatronics for my Master's feels right because honestly I've never been able to pick just one side. I love coding, I love hardware, and I love the moment when the two actually work together. That's where the fun is.
Cars and motorcycles are a big part of my life. I love engines, I love modding, and I love understanding exactly why something performs the way it does and then changing it. Getting under the hood and actually tweaking things is just as satisfying to me as writing code. I follow F1 closely too, and at some point I just built a full telemetry dashboard so I could replay races the way I actually wanted to see them. That kind of thing happens a lot with me. I get into something, I want to understand it completely, and then I want to build something around it.
Outside of tech I stay pretty active. I play sports and I'm always looking for the next physical thing to get into. There's something about being in the middle of a game where everything else just turns off and you're fully locked in. I chase that feeling a lot.
The way I approach projects is pretty similar. I like figuring out what's actually going on under the surface and iterating until it feels right. I'm not great at leaving things half finished. If something isn't working I'd rather stay with it and figure it out than move on and forget about it.
I also really enjoy collaborating. Some of my favorite projects came out of just sitting with people, throwing ideas around, and building something none of us would have come up with alone. That energy is hard to replicate and I look for it wherever I can.
Anyway, this is my corner of the internet. Have a look around and feel free to reach out.
Let's Talk
An initiative by the Practical Application of Computer Science (PACS) club in collaboration with VIRO. This Android application aims to reduce food waste by connecting event organizers with students and club members interested in receiving surplus food after events. The app facilitates a sustainable solution, ensuring that excess food doesn't go to waste and benefits those in need.
This is a personal portfolio built with Next.js to showcase my academic projects, skills, and experiences. It includes an interactive "About Me" section with an image carousel and a dynamic project listing with links to GitHub repositories and live demos.
This is a 3D animated portfolio built with Three.js and React. It includes a 3D model of me, a dynamic project listing with links to GitHub repositories and live demos, and a contact form. The 3D model can be rotated and zoomed in/out using the mouse or touch gestures on mobile devices.
TaskTonic is a task management application that allows users to create, edit, and delete tasks. It features a responsive design, a dark mode toggle, and the ability to filter tasks by their status (All, Active, Completed). The tasks are stored in the browser's local storage, so they persist even after the page is refreshed.
Minesweeper is a classic game of strategy, where the objective is to clear a grid of squares by revealing all the safe squares and avoiding revealing the bombs. The game board is a 9x9 grid, with a randomly generated number of bombs. The player can flag squares with the right mouse button to indicate they think they contain a bomb, and the game will automatically reveal the bombs when necessary.
Python Flashcards is a command-line application that helps users learn Python programming concepts through flashcards. The user can view a list of flashcards, add new flashcards, delete flashcards, and quiz themselves on the flashcards. The application uses a JSON file to store the flashcards, which allows the user to save their progress between sessions.
Chess is a JFrame application that allows two players to play a game of chess. The game board is displayed, and the players take turns making moves. The application uses object-oriented programming principles to represent the chess board, pieces, and players. The player can move pieces, capture opponents, and checkmate the opponent. Currently still under work attempting to implement ai functionality with Minimax algorithm.
PawFind is a full-stack pet shelter adoption platform built with React, Node.js/Express, and MongoDB. Features role-based access (user vs. admin), JWT authentication, and real-time support chat between users and admins powered by Socket.IO. Users can browse, filter, search, and adopt pets. Admins get a full dashboard to add, edit, and delete shelter inventory and manage live support conversations. Deployed with Docker Compose.
A full-stack Formula 1 race replay dashboard built with React and FastAPI. Uses the FastF1 Python library to pull real session data and lets you replay any race from 2018 onward, watching all 20 cars move around the track in real time. Features a live leaderboard, lap counter, per-driver telemetry (speed, throttle, brake, gear, RPM, DRS), adjustable playback speed, and a race-start detection algorithm. Deployed with a Vite/React frontend on Vercel and a FastAPI backend on Railway.
