MLOps Engineer

Role Overview

Own the infrastructure, orchestration, and scaling of machine learning systems so models move smoothly from research notebooks to high-availability production services.

Responsibilities

• Design and implement end-to-end MLOps pipelines for automated model training, validation, and deployment.
• Build and operate MLOps tooling using Kubernetes, Docker, Kubeflow, and MLflow.
• Provision and manage scalable cloud infrastructure on AWS or GCP for distributed training and low-latency inference.
• Build reliable data and feature pipelines to ensure low-latency data access and integrity across training and serving.
• Implement model monitoring to detect data drift, concept drift, and performance degradation.
• Optimize inference latency, throughput, and resource usage using quantization, pruning, and hardware acceleration.
• Collaborate with ML engineers and data scientists to set up standardized CI/CD for model artifacts and code.

Requirements

• 3–7 years of experience in MLOps, DevOps, or machine learning engineering, focused on production infrastructure.
• Deep expertise in containerization and orchestration: Docker, Kubernetes, Helm.
• Hands-on experience with ML lifecycle platforms/model registries such as MLflow, Weights & Biases, Kubeflow, or SageMaker.
• Strong proficiency in Python and Bash.
• Infrastructure-as-code experience with Terraform or CloudFormation.
• Solid understanding of CI/CD principles and automated testing for software and ML models.

Bonus

• Experience deploying and scaling LLMs and vector databases in production.