MLOps Engineer

About The Role

The role owns the infrastructure that makes ML work in production: training pipelines, model serving, monitoring, and the CI/CD systems that take models from notebook to reliable, scalable deployment.

The team partners closely with ML engineers and data scientists to eliminate friction between experimentation and production — building the automated pipelines, observability, and infrastructure standards that let models ship fast and stay reliable.

Key Responsibilities

• Design and maintain end-to-end ML pipelines for training, validation, and deployment using tools like Kubeflow, MLflow, or Airflow
• Build and scale model serving infrastructure with Docker, Kubernetes, and frameworks like KServe, Seldon, or Triton, supporting both batch and low-latency inference
• Implement CI/CD for ML systems — automated testing, model versioning, canary deployments, and rollback strategies
• Establish monitoring and observability for production models: data drift detection, latency/throughput dashboards, and automated alerting with Prometheus and Grafana
• Manage GPU compute and cloud resources across AWS, GCP, or Azure, optimizing cost and utilization for training and inference workloads
• Collaborate with data scientists to productionize models, translating experimental code into robust, tested engineering systems
• Drive platform reliability standards: SLOs, incident response, and postmortems for ML serving infrastructure

What We Are Looking For

• 3–6 years of experience in MLOps, ML engineering, or backend/platform engineering with production ML systems
• Strong Python skills plus hands-on experience with containerization and orchestration (Docker, Kubernetes)
• Experience with at least one ML pipeline/workflow tool: Kubeflow, MLflow, Airflow, or SageMaker Pipelines
• Proficiency with cloud infrastructure (AWS, GCP, or Azure), including GPU instance management and IaC tools like Terraform
• Practical experience with model serving and monitoring in production, including drift detection and performance tuning
• Bachelor's degree in Computer Science, Engineering, or equivalent practical experience
• Bonus: Experience with LLM inference infrastructure (vLLM, TensorRT-LLM, Triton), feature stores (Feast, Tecton), or streaming data systems (Kafka, Flink)