Role overview
Interested in building the foundational machine learning infrastructure for next-generation Physics AI software?
In this role, you’ll enable ML engineers and data scientists to seamlessly train, track, and deploy models by building robust, Kubernetes-based infrastructure. Responsibilities include automating training pipelines (Kubeflow), optimizing cloud infrastructure (GCP), and writing production-level code (Python, Go) with velocity. The work blends cloud-native development, distributed systems engineering, and applied AI infrastructure.
The environment is deeply technical, blending computational physics, high-performance computing, and cloud-native software development.
If you have hands-on experience building on Kubernetes, deploying open-source MLOps frameworks such as Kubeflow or Argo, and working with cloud infrastructure tools like Terraform and Docker, this could be a strong fit. Familiarity with GCP is a plus, as is a genuine interest in Physics and experience operating in a startup environment.
This is a full-time position based in the San Francisco Bay Area. Compensation is flexible depending on experience and expectations, typically ranging from $250k–$300k base plus equity.
If you’re excited about building large-scale ML infrastructure and enabling the next generation of physics-based models, we’d love to connect.
No resume required.