[Residency/Postdoc] Research Scientist - Scientific Foundation Model

The role

Materials Intelligence Lab develops Materials Foundation Models to predict properties of organic and inorganic materials. The researcher will build large-scale datasets, research AI architectures, and perform model training and evaluation.

What you will do

• Develop Materials Foundation Model for organic and inorganic materials
• Build and preprocess large-scale material datasets
• Research and apply latest architectures like Graph Neural Network and Transformer
• Build and operate large-scale training pipelines
• Analyze and implement latest AI and Scientific Machine Learning papers
• Perform model performance evaluation and benchmarking
• Write papers for international conferences and journals

Required qualifications

• Master's or PhD graduate (including expected graduates) or experienced professional in related fields
• Machine Learning research or development experience
• Experience using deep learning frameworks such as PyTorch or JAX
• Ability to read and implement latest machine learning papers
• Ability to understand and utilize latest AI methodologies such as Deep Learning, Representation Learning, and Graph Learning
• Available for full-time work for a 1 year contract

Nice to have

• Scientific Machine Learning, Geometric Deep Learning or Foundation Model research experience
• Graph Neural Networks(GNN), Transformer, Equivariant Model research experience
• Large-scale graph data processing experience
• Experience using latest MLIP frameworks such as NequIP, MACE, MatterSim, CHGNet, SevenNet
• Publication records in major conferences and journals such as ICML, NeurIPS, ICLR, AAAI and Domain Journal

About the company

Materials Intelligence Lab is a research organization that develops next-generation new materials using AI. They aim to quickly explore and design new materials by combining machine learning and simulation.