System Design Interview Assistant
RAG system indexing 11K+ chunks with hybrid retrieval (BM25 + semantic vectors), achieving 79% recall at sub-100ms latency. Reciprocal Rank Fusion and automated evaluation with grounded answers via Groq API.
> Domain Expansion: localhost:3000
Hey, I'm
I argue with LLMs for a living. Currently a Software Developer Co-op at TrendAI, and CS student at Lakehead University.
01
Stuff I built when I wasn't arguing with LLMs.
RAG system indexing 11K+ chunks with hybrid retrieval (BM25 + semantic vectors), achieving 79% recall at sub-100ms latency. Reciprocal Rank Fusion and automated evaluation with grounded answers via Groq API.
Local RAG assistant with Qdrant vector database, BGE-M3 embeddings, and Ollama LLM inference. Upload documents and get cited answers from your own knowledge base.
CNN trained on 5,000 chest X-ray images to detect pneumonia (AUC 0.91). Containerized behind a Flask REST API for real-time inference.
02
Real teams, real deadlines, real impact.
TrendAI
Building features for the Cloud One platform using Java, Jenkins, and AWS EC2
Maintair Aviation
Shipped a Flask + React flight-schedule manager on Azure — reduced errors by 40%
Code4kids
Teaching Python and web dev to 30+ learners weekly
03
Currently a Software Developer Co-op at TrendAI, building features for the Cloud One platform. Also an Honours BSc CS student at Lakehead University with a 3.7 GPA.
I've built RAG systems, trained neural networks on medical imaging data, and shipped a full-stack app at an aviation company that cut scheduling errors by 40%.
Dean's List student and Tamblyn Merit Scholar. I also volunteer at Code4kids, teaching Python to 30+ kids.
Read more about me →Interests
My inbox is lonelier than a console.log nobody reads.
Don't leave me on read like Gojo left the plot.
Say hi →