About

I'm Brad Leclerc. I build things, break things, and write about both.

Right now that means running SkeinScribe, a commercial interactive fiction platform built on the Claude API, and conducting independent research into how AI safety systems actually behave versus how they're designed to behave. The two are related — building a platform that needs predictable AI content generation is how I ended up systematically mapping the ways it's unpredictable.

My background is cross-disciplinary and mostly self-taught. I've worked in technical support at Shopify (early employee, pre-IPO), built CMS infrastructure in Python for a startup, done freelance technical writing, and spent an unreasonable amount of time studying offensive security concepts for fun. The common thread is a pattern of learning whatever a problem requires and applying it until the problem is solved or becomes a more interesting problem.

I'm based in Ottawa, Canada. I work remotely. I'm currently open to roles in AI safety research, prompt engineering, red-teaming, and technical writing — or something nobody's thought to name yet, since the field is still making up job titles as it goes.

Get in touch →