CS Specialist @ University of Toronto Scarborough (Co-op) · Minor in Astrophysics & Astronomy · Expected Oct 2027
I like building things close to the metal — neural nets in C, autonomous robots, and full-stack platforms with real-world data. I'm interested in ML systems, robotics, and the intersection of low-level programming and intelligent software.
ML Software Developer — Sunnybrook Hospital (Oct 2024 – May 2025)
Built deep learning preprocessing pipelines in PyTorch for histopathology whole-slide imaging. Cut manual data prep time by 64% and improved training efficiency by 35%.
** 1st Place — MTA Datathon** (Nov 2024)
F1 Fantasy optimization using Python trend forecasting, risk evaluation, and budget modeling across 15+ data points. Won among 50+ participants.
| Project | Stack | Highlights |
|---|---|---|
| Neural Network Engine | C, Linux | Backprop from scratch on MNIST/CIFAR-10 — ~80% accuracy, no ML libraries |
| GLOW | Node.js, Next.js, MongoDB, Docker | Full-stack Great Lakes water temperature platform with CI/CD, JWT auth, ~90% test coverage |
| Multi-Agent Navigation Engine | C | A* pathfinding + MiniMax with alpha-beta pruning (~60% state reduction at depth ≥10) |
| Robot Soccer Autonomous System | C/C++, OpenCV, EV3 | Autonomous LEGO robot with vision + blob tracking — Top 8 at RoboSoccer |
| Probabilistic Robot Localization | C/C++, Linux | Histogram/Markov localization over grid × 4 orientations under noisy sensor input |
Languages: Python · C · C++ · JavaScript · Java · SQL
ML & AI: PyTorch · TensorFlow · NumPy · pandas · Scikit-learn · LLM APIs
Robotics & Vision: OpenCV · Sensor Fusion · Localization · State Machines · Motion Planning
Web: Node.js · Express · Next.js · React · MongoDB
Infra: Docker · GitHub Actions · Git · Linux (Ubuntu) · Jira
