About

Hi, I'm Maryam, a computer scientist in Toronto working toward graduate research in trustworthy AI and ML security. I'm applying for graduate study starting Fall 2027, and this site is where I document what I'm building and learning along the way.

Background

I studied computer science (with a math minor) at the University of Toronto, and spent about a year afterward as a software engineer — mostly backend systems, data pipelines, and compliance-focused tooling.

These days, I'm working my way into research by reproducing papers I want to understand and writing up what I learn (a lot of that ends up here). Currently working on a reproduction of Label-Only Membership Inference (Choquette-Choo et al., ICML 2021), training the models and building the attacks myself, as well an eval harness reproducing Anthropic's model-written evaluations against current frontier models.

What I'm into & learning

Right now my focus is trustworthy AI and ML security — model-behavior evals, membership inference and privacy, transparency and trust. I'm drawn to the empirical, hands-on side: building things, running experiments, and poking at models to see what falls out.

On the side, a bit of quantum computing. (And French.)

Get in touch

You can reach me at mgohargani@gmail.com, or find me on GitHub and LinkedIn.