Samsung
Intern, Thermal Modeling & Simulation Engineer (AI Memory / HBM)
San Jose, California, United States · Internship · Posted Oct 27, 2026
- Education
- Pursuing a PhD (preferred) or Master's degree in Mechanical Engineering, Materials Science, Electrical Engineering, Applied Physics, or a related field; at least one academic quarter/semester remaining.
About the job
Build and validate fast thermal models and simulations of HBM stacks and packages to support design decisions for AI memory systems.
What you'll do
- Build and calibrate CFD reference models and extend compact thermal models, validating accuracy in steady-state and transient conditions.
- Improve solver performance and run workload-driven simulations to identify hotspots and assess thermal-aware policies and cooling options.
- Couple thermal models with power simulations, document methods and results, and present findings to engineering teams and lab leadership.
What they ask for
- Pursuing a PhD or Master's degree in Mechanical Engineering, Materials Science, Electrical Engineering, Applied Physics, or a related field; at least one academic quarter/semester remaining.
- Strong foundation in heat transfer, especially transient conduction in multi-layer solids and thermal contact resistance.
- Hands-on experience with numerical methods (finite-volume, finite-element, or finite-difference) and scientific programming in Python and/or MATLAB.
- Ability to critically assess simulation results, including mesh convergence, boundary conditions, and energy balance, and explain model discrepancies.
Nice to have
- Experience with Simcenter Flotherm, Ansys Icepak/Fluent, or COMSOL; compact or reduced-order thermal models, model-order reduction, or ML surrogates.
- Background in electronics packaging, 3D-IC / chiplet thermal management, advanced cooling, HPC, sparse linear solvers, or GPU programming.
Pay
$28—$68 USD hourly base pay rate
Tools
Simcenter Flotherm, Python, MATLAB, Ansys Icepak/Fluent, COMSOL
AI focus
Thermal modeling and simulation for AI memory systems, including simulations driven by AI workload traces and potential machine-learning surrogates for physical simulation.
Model-written summary · the original posting is authoritative