MLOps Engineer

Role Overview

MLOps Engineer bridging machine learning model development and production-grade software engineering. You will build and maintain automated infrastructure so data science teams can safely deploy models, monitor performance in real time, and retrain continuously.

Responsibilities

• Design and implement scalable CI/CD pipelines for ML models using GitOps principles (e.g., GitLab CI, GitHub Actions, Argo Workflows)
• Deploy and orchestrate ML workloads on Kubernetes (e.g., Kubeflow, MLflow, Seldon Core)
• Build and maintain a centralized feature store (e.g., Feast, Tecton) to keep training and serving data definitions consistent
• Set up automated monitoring, logging, and alerting for deployed models to detect data drift and concept drift (e.g., Prometheus, Grafana, ML observability tools)
• Optimize model serving for low latency and high throughput using tools like Triton Inference Server, TorchServe, or TensorFlow Serving
• Collaborate with security/compliance teams to implement access controls, model lineage tracking, and audit trails for deployed artifacts

Requirements

• 3–6 years of experience in MLOps, DevOps, or software engineering with a strong focus on ML infrastructure
• Expert-level Python
• Strong containerization experience (Docker, Kubernetes)
• Hands-on cloud infrastructure experience (AWS, GCP, or Azure)
• Infrastructure as Code with Terraform
• Proven experience building/managing ML pipelines with orchestration tools (Apache Airflow, Prefect, or Argo)
• Solid software engineering practices (unit/integration testing, version control for code and data using DVC)

Bonus

• LLMOps experience and LLM deployment
• Triton inference server optimization experience
• Vector database administration experience (Pinecone, Milvus, Qdrant)