MLOps Engineer

• Build workflows for running and managing model training and fine-tuning jobs
• Make experiments traceable through dataset, checkpoint, and model artifact versioning
• Automate model evaluation and release checks so teams can compare changes before deployment
• Build deployment and serving workflows, including safe rollout and rollback of model versions
• Improve GPU workload scheduling and utilization, with attention to performance and compute cost
• Add monitoring and troubleshooting tools for production ML systems
• Work with teams to resolve recurring production issues
• Collaborate closely with ML and Engineering teams to make experiments reproducible and model deployment and production operations reliable

• Hands-on experience building or operating infrastructure for production ML systems
• Strong Python and Linux skills
• Ability to write maintainable production code
• Practical understanding of the ML lifecycle, from experiments and evaluation to model deployment
• Experience with containers, workload orchestration, CI/CD, and infrastructure automation
• Experience with GPU workloads and measuring performance, utilization, and cost
• Experience monitoring and troubleshooting production systems
• Experience with LLM serving tools such as vLLM, SGLang, or NVIDIA Triton
• Experience with distributed or multi-GPU training
• Experience with ML tooling such as MLflow, Weights & Biases, Kubeflow, or Ray
• Experience supporting ML workloads in private cloud or on-premises environments with data governance requirements

• Competitive package aligned with talent benchmarks
• Hands-on work shaping AI capabilities for products and real-world applications
• AI-native environment with challenges across model development, evaluation, and production AI systems
• Engineering-first culture
• High autonomy and low bureaucracy
• Space to shape meaningful AI products