Computational Materials Scientist

About Atomic Tessellator

Moore's law and MLIPs bring materials science to an inflection point where simulations can demonstrate tangible utility. We're a computational materials science startup that aims to offload as much of the research process to in-silico as possible. Over the last year, our global team of four has built an infrastructure layer to support rapid, high-throughput materials testing and design. We're pushing to cover the full gamut of engineering material properties, ranging from applications in neutron irradiation, to multiscale materials, to magnetics.

We compete on speed, shipping fast is our bread and butter. Our distributed computing infra containerises common comp. mat sci operations to build workflows (e.g. supercell, sqs, relax) - this modularity lets us develop idiosyncratic capabilities on top (like neutron irradiation, MD, FEA) and chain them together with other operations we've already built. We use Python. For heavier work, we've hacked MLIPs to enable multi-GPU inference, which lets us simulate systems up to 700k atoms, and at much larger timesteps.

We're the biggest user of our own platform, and are in the process of patenting our high temperature magnet.

About the role

You'll be working at the intersection of our distributed worker architecture, and operationalising-in-code the laws of solid state physics to build out bespoke features and capabilities for frontier engineering customers. We focus on periodic structures/crystalline materials.

As most of our work is applied, you don't necessarily have to be a decorated scientist - but we're still seeking strong foundational knowledge in areas such as DFT/force field models. What's more important to us is that you can complement our pace and momentum.

• You're a strong programmer, not just limited to scripting. What we're really looking for is the builder trait of going from zero to one on a technology as if it were play.
• You can drive an LLM as both a point of productivity leverage and a research/learning tool.
• You're a generalist in some sense, and willing to expand outside of your research focus. There is potential to be an experimentalist and oversee our materials synthesis pipeline as part of our collaboration with academic institutions.

What we look for

• Authenticity: you say what you think and do what you say
• Intelligence: you can't satisfy your curiosity
• Drive: you have short-term urgency but long-term patience
• Taste: you see and appreciate what others can't see