Multiphysics Simulation Scientist

About Periodic Labs

The most important scientific discoveries of our time won’t happen in a traditional lab. We’re an AI and physical sciences company building state-of-the-art models to accelerate breakthroughs across materials, energy, and beyond. Backed by world-class investors and growing rapidly, we operate at the pace the frontier requires. Our team brings deep expertise, genuine ownership, and an insatiable drive to push the boundaries of what’s scientifically possible.

About the Role

Join a world-class team of scientists and engineers pushing the boundaries of physical R&D in a groundbreaking lab where AI and automation unlock discoveries at unprecedented speed and scale.

Periodic Labs is developing AI that can simulate physical science, verify its own predictions, and train on the full scientific method. A core part of that mission is building high-fidelity computational models of the processes happening inside our experimental systems — and making them fast enough to serve as real-time tools for AI planning, customer deliverables, and autonomous lab operation.

We are seeking an experienced Multiphysics Simulation Scientist to develop, execute, and integrate high-fidelity physical simulations of our experimental systems, linking materials processes, thermal and mechanical environments, and electrical and magnetic behavior to our AI-driven R&D pipelines. Concretely, this means work like reducing flip-chip underfill capillary flow simulations from 10–100 hours on CPU to seconds on GPU, modeling wafer warpage and thin-film deposition physics to close the loop between our lab and semiconductor customers, and building the simulation infrastructure that lets our AI reason about physical constraints during experiment planning. This person will be embedded across computational, AI, automation, and process teams and will have direct impact on both our internal discovery engine and our customer-facing deliverables.

What You’ll Do

• Build and apply multiphysics models for a wide range of phenomena relevant to our experimental systems and customer problems: thermal, mechanical, electromagnetic, plasma, fluid flow, and chemical reaction physics — often coupled. Priority problems include capillary flow and void formation in semiconductor advanced packaging, thermo-mechanical wafer stress and warpage, thin-film deposition and plasma chamber dynamics, and magnetic and superconducting material behavior.

• Accelerate simulations to operate at AI-relevant speeds. Current ANSYS/COMSOL simulations of customer problems take 10–100 hours; the goal is to get to seconds or minutes via GPU parallelization, surrogate models, or custom solver development. You will own the strategy for how we get there.

• Interface simulations with orchestration systems and data infrastructure to enable real-time digital twins and AI feedback loops. Simulated datasets you produce become training data for our AI models and reinforcement learning environments.

• Produce diverse, high-quality simulated datasets for ML training. This includes both reproducing known physical behaviors for model validation and generating synthetic data in regimes that are difficult or expensive to access experimentally.

• Embed fast simulation engines into Onnes, our AI reasoning system, so the AI can query physical constraints during experiment planning and execution — rather than treating simulation as a separate offline step.

• Collaborate with computational, AI, automation, process, and facilities teams to optimize R&D processes. Your models should inform real experimental decisions, not just academic outputs.

You Will Thrive in This Role If You Have

• A PhD or MS in Mechanical, Chemical, Materials, or Aerospace Engineering, or a closely related discipline.

• 5+ years of hands-on experience with multiphysics modeling tools — COMSOL, ANSYS, or other finite-element or finite-volume solvers — applied to real-world problems in electronics, automotive, aerospace, or chemical manufacturing. You have shipped simulations that actually influenced engineering decisions, not just published them.

• Deep understanding of coupled physical processes: heat transfer, stress and deformation, diffusion, plasma and fluid flow, electromagnetism, and the interfaces between them.

• Strong Python skills, including the ability to build workflows that connect simulation outputs to data pipelines, ML training infrastructure, and downstream analysis.

Especially Strong Candidates May Also Have

• Experience modeling thin-film deposition phenomena — PVD, PLD, CVD, ALD, plasma dynamics — and the chemical reactions and surface physics that govern them.

• Experience building GPU-accelerated or surrogate simulation approaches: custom solvers, physics-informed neural networks, neural operators, or ML-accelerated PDE solvers. The ability to make simulations fast enough to be useful in a real-time AI loop is a major differentiator.

• Familiarity with machine learning approaches relevant to multiphysics: physics-informed neural networks, operator learning, emulation, and hybrid simulation–ML pipelines.

• Multiscale modeling experience: bridging atomistic (MD, DFT), mesoscale, and continuum descriptions of physical systems.

• Comfort working across disciplines with engineers, scientists, and ML researchers — and the judgment to know when to use a fast approximate model versus when only high-fidelity simulation will do.

• Demonstrated accomplishments recognized in your field.

Mechanics

Minimum education: Bachelor’s degree or an equivalent combination of education and training or experience

Location: Our lab is located in Menlo Park and we prefer folks to be located in Menlo Park or San Francisco but can be flexible based on role

Compensation: The annual compensation range for this role is approximately $160,000–$220,000

Visa sponsorship: Yes, we sponsor visas and will do everything we can to assist in this process with our legal support.

We’re building a team of the world’s best — the scientists, engineers, and problem-solvers who don’t just follow the frontier, they define it. If you’re driven to bring AI to life in the physical world and make discoveries that have never been made before, you belong here.