Software engineer and founder building AI-powered products and production-grade systems — from multi-agent workflows and RAG pipelines to real-world applications shipped to users.
My journey began in Mechanical Engineering with a minor in Software Engineering at the University of Calgary, where I learned the art of solving complex physical problems. Today, I apply that same engineering rigor to the digital world.
I'm not just a coder (or vibe-coder); I'm a builder. I founded FitMVMT because I believe in ownership and the power of software to solve real-world inefficiencies. I also build AI-powered systems at scale — from multi-agent workflows and RAG pipelines to internal tools that buy back hundreds of hours for engineering teams.
Whether it's architecting a scalable backend, designing a fluid user interface, building autonomous agents, or defining product strategy, I thrive at the intersection of engineering, AI, and business.
A unified frontend for custom agents speaking the AG-UI protocol. Built on CopilotKit with a pluggable agent registry — add any AG-UI-compatible backend with zero UI code changes. Features streaming chat, multi-agent switching, human-in-the-loop interrupts, tool-call visualization, and Postgres-backed conversation persistence.
A two-sided marketplace connecting personal trainers with clients. Features include real-time scheduling and messaging, payment processing, delayed trainer payouts, and revenue sharing logic. It's like Fiverr but for personal training.
An AI-powered Production Engineering assistant that reclaims 100+ engineering hours per month by automating complex data retrieval and analysis. Built a multi-source RAG pipeline aggregating data across 7 enterprise systems for real-time well-performance answers, plus agent skills for schema introspection and Plotly visualization that cut time-to-insight by 40x (10 minutes → <10 seconds). A self-learning feedback loop and reasoning engine improved response accuracy by 35%.
A client-side, multi-account compound-growth net worth projector. Model your TFSA, RRSP, 401(k), Roth IRA and more side-by-side — each with its own starting balance, return rate, and contribution schedule — and see the combined picture over time. Features flexible contribution schedules with per-contribution escalation, two projection modes (fixed horizon or goal-seek to a target), inflation-adjusted views, and stacked-area and donut visualizations. No backend, no database — accounts auto-save in the browser.
An AI-powered knowledge assistant built using Retrieval-Augmented Generation (RAG). Supports uploading, parsing, and managing documents directly from the frontend, and chatting with the knowledge base across multiple sessions.
University of Calgary
4.0/4.0 GPA