MLOps Engineer

Role Overview

Own the infrastructure and deployment frameworks that make machine learning models operational at scale. You’ll build highly automated CI/CD and continuous training (CT) pipelines to ensure models are reliably served and monitored in production.

Responsibilities

• Design and maintain scalable ML infrastructure and deployment pipelines using Kubernetes, Kubeflow, and/or MLflow for automated serving and monitoring.
• Build and maintain feature stores and automated ETL/ELT data pipelines using dbt, Apache Spark, and/or Feast.
• Develop real-time and batch model serving systems using Docker, FastAPI, and Triton Inference Server with a focus on low latency and high throughput.
• Implement monitoring for model performance, data drift, and system health using Prometheus and Grafana (plus ML monitoring tools).
• Set up automated testing, versioning, and CI/CD for models and infrastructure code to enable safe, repeatable deployments.
• Collaborate with security/compliance to enforce data governance, access controls, and secure model execution across cloud environments.

Requirements

• 3–6 years of software engineering or data engineering experience, including at least 2 years in MLOps/production ML infrastructure.
• Advanced Python skills; strong shell scripting experience.
• Experience with Infrastructure as Code (Terraform) and containerization (Docker, Kubernetes).
• Hands-on ML pipeline experience on AWS (SageMaker, EKS, S3) or GCP (Vertex AI, GKE).
• CI/CD experience with GitHub Actions and/or GitLab CI.
• MLOps platform experience with MLflow, Kubeflow, Weights & Biases, and/or Prefect.
• Bachelor’s or Master’s degree in a relevant quantitative field.

Bonus

• Experience with Triton Inference Server, production fine-tuning of LLMs, or vector databases (Pinecone, Milvus).