1B Honours Computer Science Student @ University of Waterloo
A Human-in-the-Loop (HITL) data harvesting system designed to bridge the gap between human micro-decisions and personal AI agents. Features a RAG-driven backend for proprietary dataset construction and real-time behavioral alignment via LLM-annotated telemetry.
A high-performance Reinforcement Learning environment decoupling the game engine from the neural network. Features a custom Rust-based simulator utilizing Rayon for parallel execution, bypassing the Python GIL to dramatically accelerate Dueling Double DQN training throughput.
(Published on Web Store) A high-performance Chrome Extension refactored with React & Vite. Replaced standard animations with a custom 1D Value Noise algorithm to simulate organic wind patterns, rendering 1200+ procedural petals at 60 FPS.
A lightweight browser-based action game. Engineered a high-performance particle system from scratch without external engines, creating custom physics simulations for entity interactions.
A desktop productivity application built with PyQt5. Features a custom CSS-styled interface (QSS) and a persistent local JSON backend for real-time activity tracking and history management.
An autonomous agent utilizing Deep Q-Networks (DQN). Solved "Model Collapse" via Epsilon Decay strategies and accelerated training convergence by 40% using a custom Manhattan Distance reward function.
A secure, multi-user financial dashboard deployed on DigitalOcean. Features normalized database schemas, optimized SQLAlchemy queries for low latency, and dynamic financial reporting using Chart.js.