About the position The Company Mirror Physics is a New York-based AI company working on a new frontier in scientific simulation. We design intelligent systems that understand physics from first principles, providing critical acceleration for advanced technological R&D. Today, we’re building the world’s most capable AI platform for predicting experimental outcomes in chemistry and materials science, tightly coupling physical simulation with high-throughput experimental verification, to accelerate discovery in biotech, energy, manufacturing, and other domains. Backed by leading investors and scientific experts, we are expanding our research team at a pivotal moment in the field. The Opportunity The engine of the Mirror platform is our foundation model for physics-based simulation. As the lead AI researcher in physics model development, you will spearhead the development of new architectures, training algorithms, and evaluation workflows to convert large volumes of physical simulation data into highly scalable, accurate, and general-purpose predictive engines for science and industry. Responsibilities • Develop robust, scalable, and generalizable atomistic models with high fidelity across chemical domains. • Curate heterogeneous and multi-fidelity datasets into unified training corpora; develop new objectives that maximize data efficiency. • Generate novel datasets encompassing an unparalleled diversity of chemical systems consistently computed at the highest level of theory suitable for general chemistries. • Develop diagnostic tooling for model performance, failure-mode analysis, and uncertainty quantification; propose new benchmarks that stress-test predictive accuracy, physical consistency, and extrapolation. • Engineer downstream tools to enhance model accuracy and speed including model distillation and fine-tuning methods. • Engage with the AI-for-science community through publication and contributions at NeurIPS, ICML, ICLR, and other domain venues. • Mentor junior researchers and collaborate with applied science and engineering teams. Requirements • Ph.D. or M.S./B.S. with equivalent research record in Physics, Materials Science, Computer Science, or related field with a strong emphasis on machine learning and atomistic modeling. • 3+ years experience with deep learning at scale, especially equivariant GNNs, diffusion and transformer architectures. • Strong literacy in multi-scale materials modeling from the quantum-mechanical (DFT) through molecular (MD) scales • Fluency in Python plus PyTorch and familiarity with distributed training tooling (CUDA, NCCL, Slurm). • Excellent collaboration, communication, and team-working skills. • Deep commitment and passion for advancing science. Nice-to-haves • Contributions to open-source codebases, datasets, or benchmarks in computational chemistry, CFD, or continuum mechanics. • Familiarity with JAX. Benefits • Competitive salary + meaningful equity • Full health, dental, and vision benefits for you and your family • Personal fitness budget • Unlimited PTO and all national holidays Apply tot his job Apply tot his job