AI Materials Scientist – Machine Learning Interatomic Potentials / Machine Learning Force Fields_CTO Office (Taipei)

<About the Role>
We are seeking an AI for Science Materials Scientist with hands-on experience developing and applying machine learning interatomic potentials (MLIP) / machine learning force fields (MLFF). You will bridge quantum-mechanical calculations and scalable atomistic simulations, transforming emerging research into practical tools for accelerated materials discovery.

You will integrate MLIP/MLFF, DFT, generative AI, high-throughput screening, and materials science to discover and validate advanced materials, with an initial focus on magnetic materials. Experience in MLIP development, atomistic graph neural networks, computational materials science or magnetic materials is highly valued.
Strong technical communication and problem-solving skills are essential. You should be able to clearly explain your contributions to MLIP/MLFF projects, including model selection, reference-data preparation, training and validation, simulation applications, quantitative results, and how you diagnosed and resolved technical challenges.

<Key Responsibilities>

• Develop and improve MLIP/MLFF models:

Implement, train, fine-tune and adapt models for targeted material systems, making informed choices on model architecture, training strategies and computational efficiency. Develop workflows for structural optimization, molecular dynamics and selected material-property calculations.

• Perform DFT calculations and build reference datasets:

Generate and curate consistent energies, forces, stresses and other important properties from first-principles calculations, covering relevant structures, compositions, distortions, and thermally sampled configurations.

• Establish rigorous model validation:

Benchmark against DFT reference results and assess accuracy, transferability, structural relaxation, and molecular dynamics stability. Identify failure modes and define reliable application ranges.