🚀 Open to full-time as well as short-term contract roles in ML / AI / Systems Engineering

Hey 👋 I am Shadab. I’ve spent the past few years working broadly in Trustworthy AI, both during my Master’s at UBC and through research visits at a few labs across North America and Asia. Along the way, I’ve been incredibly lucky to learn from mentors and colleagues who sharpened my thinking and shaped how I approach research and engineering.

In a previous life I took a brief detour into systems biology, using ideas from formal verification to study gene regulatory networks and synthesize cell-fate determination models of C. elegans earthworm. I hope to return to similarly exciting research questions someday.

Most recently, I worked with the lovely folks at Transformer Lab, expanding their platform’s scope and polishing its UX so AI researchers can seamlessly train, evaluate, and scale models from local hardware to GPU clusters.

What I’m looking for next

I’m drawn to early- and mid-stage startups and small research teams, including those within larger organizations. I like the pace, a bit of chaos, and the satisfaction of building something that makes someone’s life easier.

I’m happiest when I can move between research and engineering. I like designing experiments to answer a question and building systems that put the answer to use. I’d also be happy in a role focused entirely on either side. What matters most to me is the problem I get to work on.

My research background in Trustworthy AI naturally draws me toward AI safety. One question I’d like to work on is which safeguards hold up once people can download and modify a model themselves. As open-weight models approach frontier performance in areas like cyber offense, I want to help evaluate those risks before release and develop safeguards that still work after release. I’m also excited about applying AI to problems in computational biology.

I’m open to other domains, too. What matters to me most is doing meaningful work with people I enjoy working with and can learn from.