MLOps Data Engineer
Hybrid 2/3 days London
Outside IR/35

Were looking for a Data Engineer with strong MLOps ownershipsomeone who builds reliable data pipelines and designs, runs, and improves ML pipelines in production. You wont be training models day-to-day like a Data Scientist; instead, youll enable Data Science by delivering high-quality datasets, reproducible training pipelines, robust deployments, and monitoring that keeps ML systems healthy and trustworthy. What youll do

• Design, build, and operate scalable data pipelines for ingestion, transformation, and distribution
• Develop and maintain ML pipelines end-to-end: data preparation, feature generation, training orchestration, packaging, deployment, and retraining
• Partner closely with Data Scientists to productionize models: standardise workflows, ensure reproducibility, and reduce time-to-production
• Build and maintain MLOps automation: CI/CD for ML, environment management, artefact handling, versioning of data/models/code
• Implement observability for ML systems: monitoring, alerting, logging, dashboards, and incident response for data + model health
• Establish best practices for data quality and ML quality: validation checks, pipeline tests, lineage, documentation, and SLAs/SLOs
• Optimise cost and performance across data processing and training workflows (e.g., Spark tuning, BigQuery optimisation, compute autoscaling)
• Ensure secure, compliant handling of data and models, including access controls, auditability, and governance practices

What makes you a great fit

• 4+ years of experience as a Data Engineer (or ML Platform / MLOps Engineer with strong DE foundations) shipping production pipelines
• Strong Python and SQL skills; ability to write maintainable, testable, production-grade code
• Solid understanding of MLOps fundamentals: model lifecycle, reproducibility, deployment patterns, and monitoring needs
• Hands-on experience with orchestration and distributed processing in a cloud environment
• Experience with data modelling and ETL/ELT patterns; ability to deliver analysis-ready datasets
• Familiarity with containerization and deployment workflows (Docker, CI/CD, basic Kubernetes/serverless concepts)
• Strong GCP experience and services such as Vertex, BigQuery, Composer, Dataproc, Cloud Run, Dataplex, Cloud Storage/or at least one major cloud provider, GCP, AWS, Azure
• Strong troubleshooting mindset: ability to debug issues across data, infra, pipelines, and deployments

Nice to have / big advantage

• Experience with ML tooling such as MLflow (tracking/registry), Vertex AI / SageMaker / Azure ML, or similar platforms
• Experience building and maintaining feature stores (e.g., Feast, Vertex Feature Store)
• Experience with data/model validation tools (e.g., Great Expectations, TensorFlow Data Validation, Evidently)
• Knowledge of model monitoring concepts: drift, data quality issues, performance degradation, bias checks, and alerting strategies
• Infrastructure-as-Code (Terraform) and secrets management / IAM best practices
• Familiarity with governance/compliance standards and audit requirements