Announcing Heaviside-1 and Atlas Fields Studio (now in Beta)learn more at Research
Back to news
Arena Physica

In the Arena: Nathan Mirman, Ph.D.

3 min read

My journey to Arena Physica started in physics grad school, where I got to study the first particle collisions coming out of the newly-constructed Large Hadron Collider. My job was to develop techniques to detect new types of particles, such as those predicted by theories of dark matter and supersymmetry. As it turns out, dark matter and supersymmetry were not found, but the LHC experiments did make the first observations of the Higgs boson.

After grad school I wanted to keep working on scientifically-motivated problems, but with more direct impact on people's lives. I spent a couple of years building things using ML in the private sector, and was eventually introduced to Pratap and the team. I joined Arena Physica at the height of Covid without ever meeting the team in person. It's been a wild ride and a pleasure to watch the company grow ever since.

Today, I'm a Research Scientist working on the EM foundation model, focusing on model training and simulation. I've been involved in a variety of different topics during my time at Arena Physica, including EE simulation in the loop, agentic frameworks for test and debug, and even a short stint trying to get LLMs to embody human personalities.

What drew me to the company is that at Arena Physica you have the opportunity to work on cutting-edge technology, and then to deploy it directly to the end user. I believe this combination of being technologically ambitious while ultimately grounded in real-world impact is one of our unique strengths. The people are pretty great too.

I'm most excited about recent advances in AI for physics and their potential to transform how hardware is designed. Today, many physics simulations require specialized solvers, expensive computing resources, and minutes or hours to produce a single result. A physics foundation model can make these calculations orders of magnitude faster and cheaper, enabling engineers to explore ideas and iterate in entirely new ways. I believe this could do for hardware design what large language models are doing for software, and I'm thrilled to help shape what comes next.

I've had the chance to be part of forward deployments on several different continents. My favorite memory is from a deployment in Buenos Aires: after a long and productive day working with our users, we found a rooftop above our Airbnb and relaxed with some cold beers in hand. Truly a moment of zen with the team.

My advice for anyone interested in working at Arena Physica is to find something you're passionate about. Be curious, go deep, and learn as much as possible from the people around you. Think critically about the world and don't take things as given. If that sounds like you, we're actively hiring for Research Scientists — you can learn more at arenaphysica.com/careers!

And in classic Arena Physica tradition, here's my hot take: free will is an illusion.