MLOps Engineer

Marlabs • San Diego, CA

109d • Hybrid

About The Position

Marlabs is hiring an MLOps Engineer to build and scale ML infrastructure supporting next-generation in silico protein design and engineering. The role bridges AI research and production systems across machine learning, computational biology, and high-performance computing. The engineer will work closely with computational scientists and platform engineers to accelerate research, bringing software, DevOps, and data engineering skills across CI/CD, orchestration, and distributed training. The position requires onsite work three days per week in Indianapolis, Indiana, or San Diego, California.

Requirements

• At least 5 years of overall industry experience in software engineering, DevOps, data engineering, or ML engineering roles.
• At least 3 years of focused MLOps experience building and maintaining production-grade ML infrastructure.
• Proven CI/CD expertise with GitHub Actions and strong DevOps practices, including infrastructure as code, version control, and collaborative workflows.
• Hands-on Kubernetes experience deploying and managing containerized ML workloads, with familiarity using container registries.
• Proficiency with SLURM or similar job schedulers in HPC environments.
• Experience optimizing distributed training, including mixed precision and checkpointing.
• Strong Python skills and experience with major ML frameworks, including PyTorch, TensorFlow, or JAX.
• Experience building ETL processes and scalable data and feature pipelines.
• Experience with cloud platforms such as AWS, GCP, or Azure.

Nice To Haves

• Experience with scientific data and protein structure formats such as PDB and mmCIF.
• Experience with protein AI models, including ESM.
• Experience with agentic systems such as MCP, LangGraph, and LangChain.
• Interest in protein language and generative models.

Responsibilities

• Build and maintain ML infrastructure, including GitHub Actions CI/CD pipelines for model training, evaluation, and deployment.
• Orchestrate compute across Kubernetes clusters and SLURM and HPC environments to optimize utilization for large-scale training.
• Develop robust, scalable data pipelines to deliver ML-ready datasets from biological sources such as PDB and mmCIF files, sequence databases, and assay readouts.
• Create tools and frameworks that enable rapid iteration on protein language models, diffusion models, and other generative approaches.
• Architect systems that scale across distributed environments and support multimodal datasets for large foundational models.
• Implement monitoring, logging, and alerting to support reliability, performance, and cost efficiency of production ML systems.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior
• Education Level: No Education Listed