Computational Materials Science Researcher (Machine Learning & Materials Informatics)

We are looking for a motivated and capable researcher in Materials Informatics and Machine Learning, with an MSc graduate/student or PhD graduate/student background in Materials Science and Engineering.

The selected candidate will design and develop machine-learning models to predict materials properties, collect and curate datasets from experimental/computational sources and literature, and translate model outputs into actionable scientific insights.

Required Qualifications

• MSc graduate/student or PhD graduate/student in Materials Science and Engineering
• Strong programming skills in Python and scientific/ML libraries (NumPy, pandas, scikit-learn, and PyTorch or TensorFlow)
• Ability to develop, debug, and document scientific code
• Hands-on experience building, training, and evaluating ML models on materials data
• Solid understanding of processing–structure–property relationships in materials
• Ability to independently design and execute research projects and analyze results

Preferred Qualifications

• Graduate or current student of Sharif University of Technology, University of Tehran, Amirkabir University of Technology, Tarbiat Modares University, or Shahid Beheshti University
• Experience with machine-learning interatomic potentials (e.g., MACE, NequIP, DeePMD, ACE)
• Experience generating data with DFT (VASP) or Molecular Dynamics (LAMMPS)
• Familiarity with graph neural networks for materials (e.g., CGCNN, ALIGNN)
• Familiarity with materials databases (Materials Project, OQMD, AFLOW) and tools such as pymatgen and matminer
• Familiarity with active learning and Bayesian optimization
• Experience with Git and working with HPC or GPU
• Publications in materials informatics or machine learning for materials

Required Documents

• CV including a list of publications, thesis, research projects, or completed ML/simulation work
• A GitHub link with sample projects and Python code; if you have no GitHub, at least one relevant work sample is required

Key Responsibilities

• Develop machine-learning models to predict structural, mechanical, thermal, electronic, and thermodynamic properties of materials
• Collect, clean, and curate materials datasets from experiments, literature, databases, and simulations
• Design featurization strategies, including compositional/structural descriptors and graph representations
• Train, validate, and benchmark models with appropriate data splits, uncertainty quantification, and avoidance of data leakage
• Train and apply machine-learning interatomic potentials and universal potentials, validated against DFT/MD
• Use active learning and Bayesian optimization to guide simulations and experiments
• Write reproducible, documented Python workflows for data pipelines, preprocessing, training, tuning, analysis, and visualization
• Interpret models (e.g., feature importance, SHAP) and translate predictions into physically meaningful conclusions
• Compare predictions with experimental data, theoretical models, and literature; contribute to reports and publications

How to Apply

• CV with a list of publications, thesis, research projects, or completed ML/simulation work
• A GitHub link with sample projects and Python code; if you have no GitHub, at least one work sample related to machine learning, data analysis, or scientific computing is required