Computer Science and Mathematics graduate from Memorial University of Newfoundland, St. John's. I build things that actually get used — my most recent project reached 2,000–3,000 HR professionals across India, the US, UK, and Mexico. That's the kind of real-world feedback loop I'm always chasing.
Right now I'm training for Ironman 70.3 and running 75 Hard at the same time. Four races on the 2026 calendar: Toronto Marathon May 3rd, a 10-miler in St. John's June 28th, Sprint Tri July 26th, and the Ironman itself December 13th.
I ranked 1st nationally in Vedic Maths and co-authored a mathematics theorem with Dr. Danny Dyer at Memorial University. I've always been the "but why does this actually work" kind of person — it turns out that helps a lot with debugging. I also carry a 🧭 compass tattooGot it before I'd been anywhere. Still wearing it everywhere. everywhere I go.
Currently looking for exciting opportunities in Software, Backend, or Data Engineering across Canada. If you have something interesting, let's talk.
Built the JD Optimizer Agent end-to-end — LLM-powered platform adopted by 2,000–3,000 HR professionals across 4 countries for job description rewrites, skill extraction, and bias detection. Architected a serverless AWS backend (Lambda, API Gateway, S3, DynamoDB) supporting multiple internal AI tools. Also delivered the Interview Prep Agent and led the Data Fellowship 2025 cohort as Squad Leader.
TA for CS courses across two years — labs, office hours, grading, the works. Explaining recursion to someone at 9pm who's had three coffees teaches you a lot about your own understanding. Best way to actually learn something is to have to teach it.
Co-authored a theorem in combinatorial mathematics with Dr. Danny Dyer. Spent a summer working through proofs, LaTeX, and the kind of problems that don't compile — they either converge or they don't. Ended up with my name on something that'll outlive any code I've ever written.
First real internship. Conversational AI company — worked on backend features, got thrown into a codebase I didn't fully understand on day one, and figured it out. Confirmed that building things people actually talk to is genuinely interesting.
Free, personalized 7-week training plan generator for the Tely 10 road race. Enter your current pace and the plan adapts your weekly mileage, long runs, and race-day targets to get you to the finish line.
Gym discovery is genuinely broken — so I built a fix. Geolocation search, real-time availability, Maps API integration. Reached 2–3k active users. Full-stack: Next.js frontend, FastAPI backend, PostgreSQL, containerized with Docker.
Built a game engine from scratch in C++ to understand how the fundamentals actually work — not just use Unity. Scene graph, physics, collision detection, sprite rendering, event system. Watch the demo.
LLM-powered platform that rewrites job descriptions, extracts skills, and detects bias. Built end-to-end at Apexon — adopted by 2,000–3,000 HR professionals across India, the US, UK, and Mexico. Serverless AWS backend: Lambda, API Gateway, S3, DynamoDB.
Cross-platform mobile app for social recipe sharing, collaborative cookbooks, and potluck planning. Built in a 4-person agile team. Owned the full recipe module — voice-to-text input, image uploads, real-time Firestore sync, Firebase Auth.
Email validation API — SMTP verification, disposable email detection, MX record lookup, bulk endpoints. Built to solve a real deliverability problem: garbage email lists that slip through without validation. Fast, accurate, production-ready.
Full-stack Netflix UI replica — responsive layout, dynamic content rows, hover interactions, and user auth. Built with vanilla HTML, CSS, and JavaScript. No framework, no shortcuts. Good exercise in DOM fundamentals.
Research conducted with Dr. Danny Dyer at Memorial University. Topics: Cops and Robbers problem on circulant graphs, Dynamic Mode Decomposition, and the Gray-Scott reaction-diffusion model. Full papers available to download directly.
Voice assistant with NLP-based intent parsing, context retention across turns, and pluggable skill modules. Handles scheduling, queries, and system commands. Built the intent layer from scratch — not a wrapper. Later adopted internally at Apexon.
Real-time face detection and recognition using deep learning — runs entirely on-device. Originally built for an automated attendance system. Accurate, fast, and works in variable lighting.
Open to Software, Backend, and Data Engineering roles in Canada. Also happy to talk training plans, math, or whatever. I reply to everything.