The Center for Artificial Intelligence in Science and Engineering (ARTISAN) at Georgia Institute of Technology is seeking a Postdoctoral Fellow or Research Scientist to lead AI/ML research on a potential DARPA-funded project within the NODES program, focused on learning protein biological function from protein motion.
This position is interdisciplinary, sitting at the intersection of machine learning, protein language/structure models, molecular dynamics, and biophysics. The successful candidate will develop dynamics-aware representations from MD trajectories and apply them to function-relevant prediction tasks (e.g., folding and binding metrics, allostery/cooperativity where applicable).
Key research focus:
- Adapting and fine-tuning protein language models and structure-aware models using molecular dynamics trajectories
- Learning representations that integrate sequence, structure, and dynamics
- Building end-to-end, reproducible ML pipelines and benchmarking against state-of-the-art methods
- Producing publishable results demonstrating how protein motion informs biological function
Required background:
- PhD in Computer Science, Bioinformatics, Computational Biology, Chemistry, Biophysics, Electrical Engineering, or a related field
- Experience in at least one of:
- Deep learning (including geometric/equivariant methods)
- Protein language or structure-based models
- Representation learning for scientific data
Preferred experience:
- Hands-on experience with PyTorch and/or JAX
- Prior work with molecular dynamics data or time-resolved molecular/structural signals
- Multimodal learning over sequence/structure/dynamics or long-context modeling
Details:
- Start: Early 2026
- Location: Atlanta, GA (can be remote initially; transition to on-site later)
- Appointment: 1-year, with possible extension based on progress and funding
- Compensation: Competitive, per Georgia Tech Postdoctoral Fellow / Research Scientist policies
How to apply:
Please email the following to
artisan...@groups.gatech.edu:
- CV (including publications)
- Brief letter of interest highlighting relevant ML and protein/MD experience
- 2–3 references