Technical
Tools for turning ideas into something real.
Queen’s University · Class of 2027
Evan.
Commerce Student, Queen’s University
(Smith School of Business)
Curious about business. Driven by data.
Learning by building things.
Business perspective. Builder’s curiosity.Get to know me
01 / About
Hello,
I’m Evan.
I’m a fourth-year Bachelor of Commerce student at Queen’s University’s Smith School of Business. I’m interested in where business, data, and product meet, and what happens when you put an idea into practice.
From researching markets to building a trading script and creating finance videos, I enjoy making complex ideas useful. Outside of that, you’ll probably find me bouldering or talking about cars.
Smith School of BusinessQueen’s University · BCom (Honours), 2027
Analytical
Finding the signal and making sense of it.
Product
Understanding people, then deciding what to build.
02 / Projects
Ideas into practice.
Things I’ve built, experiments I’m running,
and what I’m learning along the way.
Automated
Trading Bot
A Python script trading on an Interactive Brokers paper account. Live since July 2026, it’s an ongoing experiment in automation and learning how trading systems work.
Paper-trading simulation
Finance,
in short.
A YouTube channel exploring ongoing financial dilemmas and real-world scenarios through short-form videos. I research topics, make the videos, and use audience insights to improve each one.
Creating since June 2026SMALL VIDEOS. BIG QUESTIONS.
10Mviews
4Ksubscribers
Since June 2026YouTube
IC Boxscore
An NBA analytics app with transparent statistical predictions and historical player exploration. A React and TypeScript frontend backed by Cloudflare Workers and D1 makes the evidence, missing data, and limitations clear.
Open IC BoxscoreHistorical exploration works without a key. Live games need a separate free provider key; live injuries are not connected.
Behind the build and more screenshots
Transparent by design
Replaced paid language-model picks with reproducible statistical forecasts using only earlier game data. The backend uses parameterized queries, shared ingestion, bounded caching, timeouts, and capped retries.
Retrospective evaluation: 64.3% winner accuracy versus a 54.7% always-home baseline on 2,503 games in a 2023–25 evaluation. The model abstained on 31 additional games. This is a historical reconstruction, not current accuracy or betting performance; the archive has coverage gaps.
Player history explores downloaded historical game logs. Player projection accuracy has not been evaluated.
03 / Off the clock
Beyond the screen.
A few things I’m into
outside of work and school.
Bouldering
One of my favourite ways to spend time away from a desk.
Cars
Japanese domestic market cars, Mercedes, and BMW. Always happy to talk cars.
Osmow’s
A simple answer to “where should we eat?”
NBA
Following basketball and exploring how data can help explain the game.
Football
Another sport I enjoy following away from work and school.
Melodic bass
A favourite corner of electronic music.