MLOps Engineer

About The Role The role owns the infrastructure that turns machine learning experiments into reliable production systems - CI/CD for models, scalable training and serving infrastructure, and the observability layer that keeps it all running. You will work directly with ML engineers and data scientists to solve problems where reproducibility, latency, and cost efficiency matter as much as model quality. This is hands-on platform engineering for teams shipping ML at scale. Key Responsibilities Build and maintain CI/CD pipelines for ML workflows using tools like GitHub Actions, GitLab CI, or Jenkins, with automated testing and model validation gates Deploy and scale model serving infrastructure using Kubernetes, Docker, and frameworks like KServe, TorchServe, or Triton Inference Server Implement ML experiment tracking and model registry workflows using MLflow, Weights & Biases, or SageMaker Model Registry Design monitoring and alerting systems for production models: data drift detection, latency dashboards, and automated retraining triggers Optimize training and inference costs across GPU clusters (AWS, GCP, or Azure), including autoscaling policies and spot-instance strategies Automate data pipeline orchestration with Airflow, Dagster, or Prefect, ensuring reproducibility from raw data to deployed model Partner with data scientists to productionize models, translating research prototypes into versioned, observable, fault-tolerant services What We Are Looking For 3–7 years of experience in MLOps, ML engineering, or DevOps with substantial ML infrastructure ownership Strong hands-on experience with Kubernetes and Docker in production, including serving stateful ML workloads Proficiency in Python and at least one infrastructure-as-code tool (Terraform, Pulumi, or CloudFormation) Experience with cloud ML platforms and services: AWS (SageMaker, ECS, S3), GCP (Vertex AI, GKE), or Azure ML Solid grasp of ML lifecycle concepts: model versioning, feature stores, A/B testing, and drift monitoring Bachelor's degree in Computer Science, Engineering, or equivalent practical experience Bonus: experience with GPU scheduling (Slurm, Ray, Kubeflow), feature stores (Feast, Tecton), or LLM inference optimization (vLLM, TensorRT)