Staff MLOps Engineer, Apptronik

• Strong software engineering fundamentals (testing, code review, documentation, git) and proven experience in a backend or infrastructure role
• Professional experience managing cloud infrastructure on a major cloud platform (e.g., GCP, AWS, Azure)
• Hands-on experience with infrastructure-as-code tools (e.g., Terraform, Ansible, CloudFormation)
• Familiarity with ML frameworks (PyTorch, TensorFlow) and understanding of model training workflows
• Proficiency with containerization and orchestration technologies (e.g., Kubernetes, Docker)
• Understanding of distributed computing concepts and cluster management for compute-intensive workloads
• Solid understanding of Python and experience with scripting for automation and tooling
• Bachelor's or Master's degree in Computer Science, Engineering, or a related technical field
• Minimum of 4 years of professional, full-time experience building and maintaining reliable, scalable systems
• Exposure to ML/data engineering infrastructure
• Experience building tools or platforms used by other developers and researchers
• Prolonged periods of sitting at a desk and working on a computer
• Must be able to lift 15 pounds at times
• Vision to read printed materials and a computer screen
• Hearing and speech to communicate

Desirable

• We are seeking an experienced MLOps Engineer to own and maintain our cutting-edge reinforcement learning (RL) training infrastructure
• In this role, you will be responsible for the entire lifecycle of our RL systems, from managing cloud resources to optimizing job submission and deployment
• You will work closely with our AI researchers to ensure they have a stable, efficient, and scalable platform to develop and train next-generation RL models
• Design, Deploy, and Maintain Infrastructure: Manage and scale our RL training clusters on major cloud platforms (e.g., GCP, AWS, Azure) using infrastructure-as-code principles
• Orchestration and Deployment: Utilize container orchestration tools (e.g., Kubernetes, Docker Swarm) to manage the deployment and scaling of our applications and clusters
• Job Scheduling and Execution: Implement and manage tooling for submitting and monitoring large-scale distributed training jobs using modern distributed computing frameworks (e.g., Ray, Slurm)
• Database and Storage Management: Oversee our cloud-native database solutions, ensuring efficient storage and retrieval of large datasets, including images
• Developer Tools: Create SDKs, documentation, and CLI/GUI tooling that make it easy for researchers to launch experiments, visualize results, and debug issues without infrastructure expertise
• System Optimization: Implement robust monitoring, logging, and alerting to ensure the reliability, performance, health and of the training infrastructure
• CI/CD and Automation: Develop and maintain CI/CD pipelines for automated testing, data processing, benchmarking, and model experimentation
• Cross-functional Collaboration: Work closely with AI researchers and robotics engineers to understand pain points, optimize training workflows, and develop solutions that accelerate development cycles