I'm a computer vision researcher finishing my MASc at Queen's University. My work is on 6D object pose estimation: figuring out not just what an object is, but where it is and how it's oriented in 3D space.
I build full pipelines for this. Training segmentation and regression models, classical geometry methods like RANSAC, ICP, and Horn's method, and deploying the resulting inference pipelines in real-time robotic setups.
Recently I've also been working on multimodal AI, combining pose estimation with LLMs so you can ask natural language questions about a 3D scene.
- 🎓 Finishing my MASc thesis on template-based pose estimation (code will be open-sourced once it's published)
- 🔧 A multimodal pipeline that takes pose estimation output and feeds it to an LLM for scene reasoning
Core: Python, PyTorch, OpenCV, Open3D, NumPy
ML/AI: deep learning, instance segmentation, 6D pose estimation, LangChain, RAG, HuggingFace
Tools: ROS2, Docker, W&B, PyTorch Lightning, SLURM
| Project | What it is |
|---|---|
| bin-picking-pose-estimation | 6D pose estimation for robotic bin-picking. Placed in IBPC 2025, sponsored by OpenCV. |
| LawQA-RAG | Legal question answering over 100k documents, using Llama-2, LangChain, and Pinecone. |
| rag-research-assistant | RAG assistant for research papers with cross-paper citations. Built with LangChain, ChromaDB, and Gemini, deployed on HuggingFace Spaces. |
📍 Kingston, Ontario. Open to ML/AI engineering roles.
