Senior Scientist, Materials Modelling, IAIC

A*STAR - Agency for Science, Technology and Research; Singapore; On-site; SGD 140,000 - 190,000; Full time

Job summary

A*STAR IAIC invites two new Research Scientists to advance AI-driven materials modelling and discovery. You will design MLIPs, run DFT workflows, and build scalable high-throughput pipelines to explore alloys, semiconductors, and thermal materials.

You will collaborate with computational scientists and AI researchers, publish in journals, and contribute to open-source software underpinning the workflow. Strong independence and teamwork are essential.

Qualifications

• PhD in Materials Science, Physics, Chemistry, or a related discipline.
• Strong background in atomistic materials modelling, computational thermodynamics, or condensed-matter theory.
• Hands-on experience with plane-wave DFT calculations using packages (e.g. VASP, Quantum ESPRESSO).
• Hands-on experience developing, training, fine-tuning, and applying MLIPs or other atomistic ML models (e.g. M3GNet, TensorNet, MACE, GRACE).
• Experience with high-throughput materials workflows and provenance tracking.

Responsibilities

• Develop and apply foundation potentials and MLIPs to predict phase stability and thermal properties.
• Design and execute high-throughput screening workflows for multiple material properties.
• Perform plane-wave DFT calculations to generate reference data for MLIP fine-tuning.
• Develop materials property databases.
• Collaborate with computational scientists, AI researchers, and experimental teams to validate predictions.

Skills

Atomistic modelling

MLIP development

Python programming

High-throughput workflows

Education

PhD in Materials Science

PhD in Physics

Tools

VASP

Quantum ESPRESSO

Python tooling

Job description

Short Description

Research Scientist Positions in AI-Driven Materials Modelling and Discovery

Two openings for Research Scientists are available in the Sustainability Directorate at the A*STAR Institute of Advanced Intelligence and Computing (IAIC). We are seeking highly motivated researchers with expertise in atomistic modelling, machine learning interatomic potentials, and high-throughput computational workflows to develop next-generation AI-driven materials modelling and discovery capabilities. The researchers will initially apply these capabilities to thermal-management materials, while developing transferable models and workflows for broader materials discovery and engineering, including alloys and semiconductors.

Job Description

Successful candidates will conduct cutting-edge research at the intersection of first-principles simulation, machine learning, and high-throughput computational materials design. The work links three scales: density functional theory (DFT) ground truth, foundation-potential and fine-tuned machine learning interatomic potential (MLIP) simulation of finite-temperature transport and mechanical properties, and continuum package-level thermal and thermomechanical modelling driven directly by the computed bulk properties.

Key Responsibilities Include

• Develop and apply foundation potentials and fine-tuned MLIPs to predict phase stability, lattice and electronic thermal conductivity, thermal expansion, and elastic properties at scale.
• Design and execute high-throughput screening workflows on thermal expansion, thermal conductivity, and stability, among other properties.
• Perform plane-wave DFT calculations to generate reference potential-energy-surface and property data for MLIP fine-tuning and validation.
• Develop materials property databases.
• Collaborate closely with computational scientists, AI researchers, and the experimental teams to validate predictions, interpret characterisation data, and generate new hypotheses.
• Engage with industry partners and relevant stakeholders, including co-authoring industry-facing technical briefs.
• Publish research findings in leading international journals, present at major scientific conferences, and contribute to software underpinning the workflow.

The candidates should be capable of working independently while contributing effectively within a multidisciplinary research team. Strong analytical thinking, scientific curiosity, problem-solving ability, and excellent written and verbal communication skills are essential.

Technical Requirements

• Ph.D. in Materials Science, Physics, Chemistry, Chemical Engineering, Mechanical Engineering, or a related discipline.
• Strong background in atomistic materials modelling, computational thermodynamics, or condensed-matter theory.
• Hands-on experience with plane-wave DFT calculations using established electronic structure packages (e.g. VASP, Quantum ESPRESSO)
• Hands-on experience developing, training, fine-tuning, and applying MLIPs or other atomistic machine learning models (e.g. M3GNet, TensorNet, MACE, GRACE, etc.).
• Experience with molecular dynamics and lattice-dynamics methods for thermal and mechanical property prediction, such as Green-Kubo or non-equilibrium MD, phonon Boltzmann transport, and quasi-harmonic approximations.
• Experience in high-throughput materials workflow automation, including automated error handling and provenance tracking.
• Strong programming skills in Python and familiarity with modern scientific software engineering practice (version control, testing, packaging). Knowledge of the pymatgen / MatGL / MatCalc ecosystem is a strong advantage.
• Experience with high-performance computing environments, including GPU-accelerated workloads.
• Experience with the application of agentic AI in orchestration and coding is a major plus.

Experience In The Following Will Be Viewed Favourably

• Continuum thermal and thermomechanical finite-element modelling (e.g. COMSOL, ANSYS) and coupling of atomistic-derived properties into continuum models.
• Alloy design, short-range order, and multiple-principal-element alloys.
• Prior contributions to open-source computational materials science software.

Appointment

The positions are offered at IAIC, A*STAR. Remuneration will commensurate with qualifications and experience. Only shortlisted candidates will be notified.