Education
I hold a BSc with Honors in Physics from the University of Alberta, and an MSc in Computing Science with a specialization in Statistical Machine Learning from the University of Alberta. My MSc thesis (with Czaba Szepesvári) was on the topic of pure exploration problems in multi-armed bandits. I did a research internship with the Advanced Concepts Team at the European Space Agency, where I used machine learning to analyze chaotic trajectory data and rode my bike on endless Dutch bike paths with childlike glee.
About Me
I’m a machine learning engineer at Atomic AI, where I split my time between developing machine learning models, building infrastructure to support the development and deployment of said models, and working with scientists to leverage the tools that we’re building to accelerate RNA drug discovery. We recently released the ATOM-1 whitepaper where we talk about some of the cool stuff that we’ve been working on.
Born, raised and educated in Edmonton, Alberta, I’m currently living in San Francisco, California. I miss outdoor hockey rinks but it turns out that quad roller skating is a pretty good substitute in the Bay Area.
In a previous life, I thought that I wanted to be a physicist, but I’ve since realized that I’m broadly interested in working on cool problems with cool people. Over the years I’ve performed my best impression of “too interdisciplinary to function”, which is likely a suboptimal approach for the “paper mill” strategy to career growth but one which I’ve found extremely fulfilling. I’ve been lucky enough to work on a wide variety of problems in a wide variety of fields; I’ve worked on problems in computational physics, computer vision, robotics, space science and machine learning, both on the theory and applied side.
I’m particularly interested in working on problems that are at the intersection of machine learning and the natural sciences, especially in areas where there is an opportunity to make a positive impact on the world, and I’m especially excited about the flurry of activity in structural biology and materials science that we’re seeing right now. I’m particularly interested in finding ways to combine the current strengths of machine-learned models with the power of traditional computational methods to achieve breakthroughs in accuracy and efficiency. For example, combining Boltzmann generators and or all-atom structure prediction models in conjunction with docking and free energy perturbation methods to accelerate drug discovery.
You can find an overview of some projects that I’ve worked on in the past here, or use the navigation bar at the top of the page. If you’d like to get in contact, my personal email is on my Github profile (also in the nav. bar), along with my LinkedIn.