From stage to systems
The pivot point
In 2015, I was a professional dancer with the National Ballet of Canada — a 12-year career spanning national and international productions. Then I herniated two discs in my lower spine. After six months of recovery and a brief comeback attempt, the injury returned. The company offered a generous severance, and my performing career was over.
What followed were years of depression, isolation, and rebuilding. I lost a 12-year relationship, sold property, cashed out savings. I was stuck in a way that felt permanent. The only direction I could see was school — any school, any direction other than where I was facing.
Building things
Long before I thought about tech as a career, I was obsessed with building computers as therapy. When my mind would race, I'd build something — Hackintoshes, custom PCs, wire sculptures. I was fascinated by subverting Apple's hardware restrictions: why pay 300% more for components just because they're in a white case with a logo?
I built Hackintoshes for myself and others, turning $500 in parts into machines that outperformed $3000 iMacs. I learned to debug hardware conflicts, troubleshoot software, and fix what broke. This stuff is massively overpriced, and people pay way too much by relying on someone else to do something they can easily learn.
The bridge to AI
I needed work, so I started doing AI data annotation — evaluating AI-generated images, labeling datasets, assessing LLM outputs. I didn't plan to understand AI better, but I did. I stopped seeing these models as perfect systems and started understanding them as highly accurate prediction engines built on imperfect human data.
In conversations with various models about career direction, they consistently pointed toward creative technology. I had artistic background, hardware skills, and I was experimenting with ComfyUI. The suggestion made sense, but I saw a problem: no professional experience, no formal education, and I can't write code from scratch.
Learning to direct, not type
Yet somehow I was building working utilities — and faster than people who can code would build them — by directing LLMs. I'd specify the logic, get an AI assist on implementation, then debug and iterate until it worked. I discovered that using models with different strengths and blind spots made a massive difference. One model alone would give me broken code. Multiple perspectives got me working systems quickly.
That's when I built MiniBook — a dual-LLM pipeline that turns a single prompt into a finished, illustrated storybook. I designed the architecture, specified the logic, and directed the implementation. The engineering judgment is mine. The method is AI-assisted. And it ships.
What I bring
Twelve years of ballet taught me that great performance is precise structure in service of making someone feel something. Hundreds of variables executing flawlessly, no rollback once the curtain rises. I learned the same lesson building Hackintoshes and debugging hardware: understand a system deeply, then make it produce something both beautiful and reliable.
I'm looking for roles where I can use that same instinct — creative technology, AI tooling, systems work where precision and judgment matter more than typing speed. I know what it feels like to contribute something meaningful to work I believe in. That's what I'm looking for again.
Open to roles in creative technology, AI tooling, and IT support. Toronto or remote.