Founding Engineer (ML Systems) – Godela

What you will be doing

• Developing scalable approaches to train high-accuracy, large-scale physics-informed models.
• Designing and testing new architectures for multi-physics and multi-scale modeling.
• Optimizing training efficiency through distributed computing, GPU optimization, and performance tuning.

Must-have criteria

• Strong Python + PyTorch (or JAX/TF) skills.
• Proven experience delivering ML systems into production.
• Experience with multi-GPU / multi-node training.
• Experience training and deploying large-scale ML models with GPU acceleration and distributed workloads.
• Hands-on AWS experience (compute, storage, IAM, cuda) or other cloud infra provider.
• Background in building or scaling training pipelines, APIs, or ML infra that supports real-world products.

Nice-to-have criteria

• Exposure to engineering simulations (CFD, FEM, PDEs) or physics-informed ML (PINNs, neural operators).
• Familiarity with Docker/Kubernetes, CI/CD, Slurm, or other workload managers.
• Experience with Graph Neural Networks, Physics-Informed Neural Networks, Neural Operators, and Transformer architectures.
• Experience working with messy, high-dimensional data.