Machine Learning Researcher - Generative Simulation Models

Machine Learning Researcher – Generative Simulation Models

Be part of a team redefining how molecules and materials are discovered through AI. This group is building large‑scale models that capture the fundamental rules of chemistry and physics, enabling predictive and generative design at atomic resolution. The work spans deep generative architectures, probabilistic reasoning, and simulation‑driven learning integrated with massive compute and data pipelines. You'll collaborate with leading experts across machine learning, physical sciences, and engineering to create systems that not only model reality but accelerate experimental validation and scale through synthetic data.

This is an opportunity to take ownership from concept to deployment and push the boundaries where advanced AI meets the structure of matter.

• Architect and train advanced generative models for molecular and materials design, leveraging approaches such as diffusion processes, autoregressive transformers, flow‑based systems, and latent‑variable frameworks.
• Develop rich structural representations for molecules and atomic systems using cutting‑edge techniques like symmetry‑aware graph networks, geometric attention models, and latent encoders that respect physical constraints.
• Create novel sampling and simulation strategies that fuse probabilistic inference, deep learning, and reinforcement learning to efficiently explore complex energy landscapes and accelerate computational discovery.

• PhD. or equivalent research experience in machine learning, physics, chemistry, computer science, or a closely related discipline.
• Demonstrated expertise in generative modeling, including diffusion‑based methods, variational approaches, normalizing flows, or autoregressive architectures.
• Hands‑on experience in representation learning for structured or geometric data, particularly graph‑based or 3D models (e.g., equivariant networks, geometric transformers).