ML Ops Engineer

Luminary Cloud • San Mateo, CA

112d • Onsite

About The Position

Luminary Cloud is transforming how the world's most innovative companies generate vast amounts of CFD simulation data for Physics AI, design exploration, and optimization. Backed by Sutter Hill Ventures, our Series B startup is at the forefront of the transition to Physics-based AI through our scalable cloud platform.

Requirements

• Bachelor's degree or higher in Computer Science, Data Science, Statistics, Applied Mathematics, or related fields.
• 5+ years of industry experience in machine learning operations (MLOps), including model development, deployment, monitoring, and scaling ML systems in production environments.
• Proficiency in Python with demonstrated experience writing production-level code. Familiarity with BASH and SQL required.
• Solid experience with Google Cloud Platform (GCP), Docker, Kubernetes, and Vertex AI for cloud-based ML infrastructure.
• Hands-on experience with modern MLOps frameworks including Kubeflow, MLflow, and Argo Workflows.
• Strong experience building scalable data pipelines, particularly with large volumes of unstructured data and familiarity with VTK for 3D geometric data.
• Proven ability to independently deploy ML models and set up inference pipelines with monitoring capabilities.
• Experience managing model registries and integrating them with automated workflows.
• Strong problem-solving skills and ability to troubleshoot complex distributed systems.

Nice To Haves

• Physics Interest: Heavy interest in simulation technology and/or High-Performance Computing (HPC) with some exposure preferred. Understanding of how ML applies to physics simulations and scientific computing is valuable.
• MLOps at Scale: Demonstrated experience building and maintaining MLOps infrastructure at scale, with focus on reliability and performance in production environments.
• Technical Excellence: Hands-on approach with ability to write code with velocity while maintaining high quality standards. Experience with additional programming languages such as Go and C++ is a plus.
• Startup Environment: Curious and quick learner who thrives in a fast-paced environment. Clear communicator with a collaborative approach to working with diverse technical teams.
• In-Office Commitment: Enthusiastic about being in-office 5 days a week, contributing to our hands-on, collaborative engineering culture.
• Infrastructure Focus: Passionate about building systems that enable other engineers and scientists to be more productive with an understanding that great infrastructure should be invisible to end users.

Responsibilities

• Build and maintain robust MLOps infrastructure enabling ML engineers and data scientists to train, track, and deploy models seamlessly without managing low-level Kubernetes infrastructure.
• Design and implement automated training pipelines and experiment tracking systems using modern MLOps frameworks including Kubeflow, MLflow, and Argo Workflows.
• Develop scalable data pipelines for large volumes of unstructured data, with particular focus on 3D geometric data using VTK and physics simulation outputs.
• Deploy machine learning models and set up production inference pipelines with focus on performance and reliability.
• Manage model registries and integrate them with automated workflows for seamless model lifecycle management.
• Implement comprehensive monitoring systems for continuous performance tracking of ML models in production environments.
• Collaborate with cross-functional teams to ensure MLOps infrastructure meets the evolving needs of physics-based AI applications.
• Write production-level code with velocity, maintaining high standards for performance and scalability.
• Optimize cloud infrastructure on Google Cloud Platform, leveraging Docker, Kubernetes, and Vertex AI for efficient resource utilization.

What This Job Offers

• Job Type: Full-time
• Career Level: Mid Level