I design and build transformer-based models, foundation models, and retrieval-augmented AI systems. My work focuses on representation learning, efficient model adaptation, and scalable generative AI, from structured scientific data to language.
Machine Learning Scientist with a PhD in Computer Science and over five years of research experience in deep learning, transformers, and large-scale representation learning. I specialize in designing and training neural architectures for complex data, from foundation models to retrieval-based AI systems.
My work spans generative AI, LLM systems, and scientific machine learning. I build complete ML systems, including model architectures, training pipelines, retrieval systems, and production tooling using PyTorch, JAX/Flax, Hugging Face, Docker, and HPC infrastructure.
Currently at York University, my research explores two directions: building world models for complex systems and developing efficient approaches for adapting foundation models using techniques such as LoRA fine-tuning.
A local RAG pipeline for automated contract compliance review, built with LangChain, Chroma, and Ollama/Llama 3, async job architecture with Pydantic-validated, structured output.
A lightweight local RAG system for codebase Q&A using LangChain and Chroma with MiniLM embeddings, plus automated pull request description generation from diffs.
A modular Transformer implemented from scratch in JAX/Flax, custom relative positional encoding, encoder block abstractions, and full documentation on ReadTheDocs.
A React/Vite TV tracking web app using Dexie.js for local storage and the TMDb API for show data, with an import pipeline for migrating watch history from TV Time.
Sixty-one chapters tracing what happens between a typed prompt and a generated response, covering tokenization, embeddings, attention, context construction, retrieval augmentation, and modern LLM systems.