About this role
PhysicsX is seeking a Principal ML Infrastructure Engineer to build and operate the AI infrastructure for training, fine-tuning, and serving large physics models. You will collaborate with research scientists and ML engineers to ensure efficient and reliable model training at scale, optimizing pipelines for performance and cost.
What we are looking for 6
- Design and operate distributed training infrastructure for neural operator architectures
- Optimize training pipelines for throughput, fault tolerance, and cost efficiency
- Build and maintain experiment tracking and observability systems
- Solve data loading bottlenecks and optimize data pipelines for efficient I/O
- Build serving infrastructure for pre-trained LPMs and design model packaging pipelines
- Improve developer experience with fast iteration cycles, reliable CI/CD, and debugging tools