Welcome! I'm a software engineer with a background in computer and mechatronics engineering, passionate about building intelligent systems that bridge the gap between software and the real world. My work spans C++, Python, machine learning, and embedded systems, with a focus on creating efficient, reliable, and impactful solutions.
Using transfer learning and knowledge distillation to train a lightweight model for mimicking a pre-trained larger model in a clinical histopathology dataset, MHIST
A multithreaded C++ stock trading platform featuring a robust market simulator. Utilizes the Factory Pattern for flexible order creation and the Strategy Pattern to enable dynamic trader behaviors for both market and limit orders, providing a realistic and extensible trading environment.
Real-time traffic light detection using YOLOv5 and HSV color classification, with overlaid driving decisions and labeled video output.
A digital travel diary and photo album platform for documenting journeys, sharing stories, and organizing memories with interactive maps and galleries.
A real-time AI-powered system for predicting flashover events in firefighting scenarios using computer vision and sensor data fusion.
An autonomous search and rescue robot capable of navigating diverse terrains using advanced sensor fusion and AI-driven path planning.
An autonomous robot designed to follow a visual line path using sensors and real-time feedback control algorithms.
A convolutional neural network (CNN) for animal classification using the LHI-Animal-Faces dataset. The model distinguishes between 19 animal classes and one human faces class, enabling robust multi-class image recognition.
A performance comparison of OpenMP (CPU) and CUDA (GPU) in accelerating Hermite N-body simulations, highlighting speed, scalability, and efficiency in modeling gravitational systems.
This project investigates the application of deep learning to Optical Music Recognition (OMR), focusing on identifying handwritten musical symbols from images. By evaluating and fine-tuning top-performing CNNs, the study aims to improve symbol classification accuracy on the HOMUS dataset. Ensemble models are also explored to enhance performance beyond individual architectures.
A smart scoring application designed to automate and enhance the evaluation process for sports and academic competitions.
An AI-powered restaurant recommendation tool that combines Retrieval-Augmented Generation (RAG) with local AI Agent to deliver smart, location-based dining insights.
University of Waterloo
University of Toronto
Emphasis in Computer Engineering, Data Analytics and Machine Learning
Feel free to add me on LinkedIn or shoot me an email. I'd love to chat!
Send me any message and I'll get back to you.
