MLOps Engineer

Overview

The MLOps Engineer will be responsible for building and managing end-to-end machine learning platforms, focusing on the CI/CD process for model training, testing, and deployment. This role requires collaboration with various teams to ensure effective model serving, monitoring, and infrastructure scaling. Contractors will work on innovative projects that are outside the scope of IR35, with options for remote and hybrid arrangements in the UK.

Responsibilities

  • Build and run end-to-end machine learning platforms.
  • Implement CI/CD processes for training, testing, and deploying models.
  • Manage model serving, monitoring, drift detection, and automated retraining.
  • Optimize infrastructure for scaling and cost efficiency.

Requirements

  • Commercial experience in an MLOps or ML Platform Engineering role.
  • Proficiency in Kubernetes (EKS, GKE, or AKS) and Docker.
  • Experience with Infrastructure as Code tools such as Terraform and Helm.
  • Familiarity with CI/CD tools including GitHub Actions, GitLab CI, Jenkins, or ArgoCD.
  • Knowledge of ML lifecycle tooling like MLflow, Kubeflow, SageMaker, or Vertex AI.
  • Experience in orchestration tools such as Airflow, Prefect, Dagster, or Argo Workflows.
  • Understanding of model serving frameworks like KServe, Seldon, Triton, BentoML, or Ray Serve.
  • Proficient in monitoring tools such as Prometheus, Grafana, Evidently or WhyLabs.
  • Strong programming skills in Python and Bash.
  • Deep hands-on experience with AWS, GCP, or Azure.