periodic-labs
Computational Scientist, Differentiable Physics
$250K–$350K
Menlo Park, CA · On-site · Full-time · Posted Sep 4, 2026
- Education
- PhD or equivalent research experience; minimum education: Bachelor's degree or similar experience.
- Visa sponsorship
- Yes, we sponsor visas and will do everything we can to assist in this process.
About the job
Build differentiable, accelerator-ready continuum-physics simulations and combine them with deep learning for scientific and engineering problems.
What you'll do
- Build and extend differentiable solvers, implement numerical methods, and diagnose convergence, stability, and modeling failures.
- Combine simulation and deep learning, then validate models against experiments, benchmarks, or high-fidelity simulations.
- Create datasets and evaluations to help LLMs accelerate and automate simulation tasks.
What they ask for
- PhD or equivalent research experience in a related field.
- Code-level experience building or substantially modifying PDE solvers, numerical methods, or differentiable simulations.
- Deep expertise in at least one continuum domain, including some fluid-dynamics experience.
- Meaningful experience building, training, and evaluating deep-learning models for physical systems.
Nice to have
- Experience with adjoint methods, implicit differentiation, differentiable programming, or scientific optimization.
- Experience accelerating scientific software on GPUs or TPUs.
Pay
$250,000-350,000 USD base salary
Tools
JAX, PyTorch, Julia, C++, Python
AI focus
Use deep learning with simulation for surrogate modeling, learned physics, inverse problems, parameter estimation, and optimization; create datasets and evaluations to guide LLM development for accelerating and automating these tasks.
Model-written summary · the original posting is authoritative