Multiphysics Simulation Scientist, Semiconductors

Periodic Labs•Menlo Park, CA

21d•$160,000 - $220,000•Hybrid

About The Position

Periodic Labs is developing AI systems that can simulate physical science, verify 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 and customer-facing systems, then making those models fast enough to support AI planning, autonomous lab operation, and engineering decision-making. We are looking for a Multiphysics Simulation Scientist to develop, validate, accelerate, and integrate models of semiconductor-relevant physical processes. This role is designed for someone with deep experience in multiphysics simulation, especially in domains such as advanced packaging, wafer mechanics, thin-film deposition, plasma processing, thermal transport, stress and warpage, capillary flow, electromagnetics, or related manufacturing processes. You will work across computational science, AI, automation, process engineering, and customer-facing teams. Your models will help us understand experimental systems, generate training data, guide process decisions, and deliver high-value engineering insights to semiconductor customers.

Requirements

A PhD, MS, or equivalent experience in mechanical engineering, chemical engineering, materials science, electrical engineering, applied physics, aerospace engineering, or a closely related discipline.
Significant hands-on experience with multiphysics modeling tools such as COMSOL, ANSYS, Abaqus, Fluent, OpenFOAM, Sentaurus, Lumerical, MOOSE, or other finite-element, finite-volume, particle, or continuum solvers.
Deep understanding of coupled physical processes relevant to semiconductor or advanced manufacturing systems, such as heat transfer, stress and deformation, capillary flow, diffusion, plasma dynamics, electromagnetics, surface reactions, phase change, deposition, or materials evolution.
Experience building simulations that influenced real engineering or scientific decisions. You have not only published or run models; you have used them to explain failures, guide designs, improve processes, or support customer deliverables.
Strong Python skills and the ability to connect simulation outputs to analysis workflows, data pipelines, ML training infrastructure, and downstream decision-making systems.
Comfort working across disciplines with process engineers, experimental scientists, ML researchers, automation engineers, and external technical stakeholders.
Good judgment about simulation fidelity. You know when a commercial multiphysics package is the right answer, when a custom model is needed, and when a fast approximation is more useful than a slow high-fidelity model.

Nice To Haves

Deep knowledge of semiconductor advanced packaging, including underfill, flip-chip assembly, thermal-mechanical reliability, warpage, void formation, interconnects, or packaging materials.
Hands-on modeling of thin-film deposition processes: PVD, PLD, CVD, ALD, sputtering, evaporation, epitaxy, or related surface and chamber dynamics.
Fluency in plasma physics, including sheath dynamics, charged species transport, reactive flows, or plasma-enhanced deposition and etching.
A track record with wafer-scale mechanics: stress, bow, warpage, thermal cycling, film stress, delamination, fracture, or reliability modeling.
Ability to build GPU-accelerated solvers, reduced-order models, surrogate models, physics-informed neural networks, neural operators, or ML-accelerated PDE solvers.
Comfort with the full quantitative toolkit: parameter estimation, design of experiments, model calibration, sensitivity analysis, or uncertainty quantification.
Skill bridging length scales, from DFT and MD through kinetic Monte Carlo, phase-field modeling, and continuum mechanics.
Familiarity with semiconductor process integration, metrology, failure analysis, process control, or customer-facing engineering workflows.
A record of recognized impact: high-citation publications, deployed engineering models, patents, major customer-facing technical contributions, or simulation tools others actually use.

Responsibilities

Build and apply multiphysics models for semiconductor-relevant systems, including thermal, mechanical, fluid, electromagnetic, plasma, chemical reaction, and materials processes, often in coupled settings.
Model priority problems such as flip-chip underfill capillary flow and void formation, thermo-mechanical wafer stress and warpage, thin-film deposition, plasma chamber behavior, thermal budgets, process-induced deformation, magnetic or superconducting materials behavior, and other customer-driven physical systems.
Use high-fidelity simulation tools such as COMSOL, ANSYS, Abaqus, Fluent, Lumerical, Sentaurus, OpenFOAM, MOOSE, or comparable platforms where appropriate, while also helping decide when custom solvers, reduced-order models, or surrogate models are needed.
Validate models against experimental and process data. You will work with experimentalists and engineers to compare simulations against measurements, estimate uncertain parameters, understand failure modes, and decide when a model is ready to guide real decisions.
Generate physically meaningful simulated datasets for ML training. Your simulations will help train AI systems in regimes where experiments are expensive, slow, or difficult to run.
Integrate simulation workflows with Periodic Labs’ AI, data, and orchestration infrastructure. Your models should become callable tools for AI planning and experiment interpretation, not standalone reports.
Collaborate with process, automation, AI, facilities, and customer-facing teams to optimize R&D workflows and solve practical engineering problems.
Help define the long-term multiphysics modeling roadmap for Periodic Labs’ semiconductor and materials programs.