Internship - MLFF Distillation & GCMC Integration
Core
Develop fast, accurate machine learning force fields (MLFFs) for high-throughput Monte Carlo simulations and integrate them into the kUPS simulation framework to evaluate next-generation MOFs.
Role type
Engineering intern (MLFF distillation & GCMC integration)
Builds
Lightweight MLFF student potentials optimized for Monte Carlo inner loops
Domain
Materials science, computational chemistry, AI for science
Deliverable
production ML models
Required skills
molecular simulation methods (GCMC, MD), adsorption modelling at atomic scale, Linux environment management, MLFF distillation, dataset curation and versioning, pipeline profiling and optimization, validation campaign execution
Preferred skills
knowledge distillation for scientific ML, active learning workflows, DFT data generation, gas adsorption/MOFs background, classical force fields expertise, published research in top-tier venues
Technologies
kUPS, GCMC, MD, Linux
Responsibilities
Distill state-of-the-art equivariant models into lightweight student potentials; curate and document training/validation datasets; run head-to-head validation campaigns against classical force fields; profile and optimize the pipeline for throughput; collaborate with computational chemists on reference data and validation strategy; contribute to a publication on MLFF-driven GCMC for MOF screening
Seniority
Intern