I’m an AI Engineer and ML Scientist / Researcher based in Toronto, focused on LLMs, agentic AI, and efficient machine learning systems.
I hold an M.A.Sc. in Electrical & Computer Engineering from the University of Toronto, where my research involved collaboration with Meta. Previously, I worked as an AI Engineer at ArcTrade, building AI tools for software review, testing, and documentation.
I approach problems from the user’s perspective: define success, establish a baseline, then build and evaluate. I bring that same discipline to AI systems, with an emphasis on reliability, practical deployment, and measurable impact.
- LLM applications and AI agents: Systems for developer workflows, academic outreach, and news briefing.
- Reliable AI workflows: Tools and evaluation processes that help assess model behavior and improve software quality.
- Efficient ML systems: Model design, optimization, and evaluation across GPU, edge, and FPGA deployment settings.
-
Scholar Link: An agentic academic outreach platform that helps students find matching professors and automate personalized emails and replies. Built with Django, PostgreSQL, and LLMs.
-
Briefora: An agentic news briefing project that uses LLMs to turn news into concise summaries and help readers understand key developments.
-
Spatons: Compact neural networks for XR display predistortion, combining model design with compute, memory, and deployment evaluation.
-
Football Player Tracking: Player detection and segmentation using YOLOv8, U-Net/U²-Net, and SAM, including jersey analysis and zero-shot mask filtering.
- Artificial Intelligence: Implementations exploring search, optimization, reinforcement learning, and deep learning.
- Sentiment Analysis Project: NLP and sentiment classification.
- Reinforcement Learning Based Fault Tolerant Scheduling: Exploration of reinforcement learning for scheduling.
- Stock Prediction Using ML: Experiments in stock and currency prediction.
- Flight Management System: A software project using Python, PostgreSQL, and Java.
LLM systems: LiteLLM, AWS Bedrock, OpenRouter, Hugging Face
Machine learning: Python, PyTorch, scikit-learn, OpenCV
Software & systems: C++, C#/.NET, SQL, CUDA, Django, PostgreSQL, Git, Linux, AWS