ML Engineer

Accepting applications until:

Job Description

Machine Learning Engineer

Global’s central Data Science function is recruiting a Machine Learning Engineer to work across our data and product portfolio.

As a Machine Learning Engineer at Global, you’ll be part of the engineering bridge between data science and production-taking models built by our data scientists and making them robust, scalable and maintainable across audience targeting, advertising measurement and content intelligence. You’ll work with colleagues across Data Science, Machine Learning Engineering and Data Engineering, bringing your own ideas to how we build and improve our models. The role is based in central London (Holborn, with occasional travel to Leicester Square).

Key Responsibilities

• Model Productionisation, Serving & Lifecycle Management (70%): Take models built by our data scientists and turn them into reliable, testable production systems; build the serving infrastructure to deliver them across batch and real-time use cases; build and maintain consistent feature pipelines; keep models healthy once live through monitoring, drift detection and periodic retraining; and help audit and migrate existing models to modern standards without disrupting live products.

• Standards & Enablement (15%): Apply and contribute to the engineering standards for how models are built, tested and deployed, aligned to the MLOps platform; help create reusable templates and documentation that enable data scientists to work independently; and feed your experience back to help shape the platform direction.

• Cross-functional Partnership (15%): Work with Data Science, MLOps, Data Engineering and Product to understand new products early and ensure models are handed off in a deployable, maintainable form.

What You’ll Love About This Role

• Think Big: This is a true AI and data-driven space-the models you ship influence what millions of listeners hear and how brands invest their media budgets.

• Own It: You’ll take ownership of getting models into production and keeping them healthy, and your voice will count in how we build going forward-not just maintaining what exists.

• Keep it Simple: You’ll build on and contribute to reusable patterns and templates rather than one-off solutions.

• Better Together: You’ll work with Data Science, MLOps, Data Engineering and Product, with plenty of opportunity to broaden your experience across Global’s data products.

What Success Looks Like

In your first few months, you’ll have:

• Built a working understanding of Global’s engineering and data science tools and how they connect.

• Established strong working relationships with Data Science, MLOps and the wider engineering team.

• Contributed to the audit of existing ML models in production, assessing stability, maintainability and risk.

• Taken a model (or part of one) to a more robust, documented state, and contributed reusable components others can use.

What You’ll Need

• Production ML experience: You’ve delivered ML projects in a commercial environment, contributing to deployment, CI/CD and lifecycle management. You’ve delivered models across different mediums, such as batch models running on a schedule and models served via APIs.

• Strong Python: Solid Python written to production standards: you unit test your code, use version control and code review as a matter of course, and code defensively so that simple edge cases don’t break production environments. Experience with PyTorch or similar ML frameworks is a plus.

• Model evaluation & monitoring: You can investigate why models underperform across data, features and architecture, and explain the trade-offs behind your choices. You understand how to monitor models in production, detect model and data drift, and put safeguards in place to defend against it.

• Orchestration: Familiarity with workflow orchestration tools such as Airflow or similar, and an understanding of how to schedule, monitor and maintain dependable data and model pipelines.

• Real-time ML & reproducibility: A good understanding of production inference patterns and reproducible environments (Docker, MLflow or equivalent), and a willingness to deepen it.

• Cloud & data tooling: Familiarity with a SQL data warehouse-style system, and experience with cloud platforms such as AWS, GCP or Azure. Experience with AWS, including SageMaker, Snowflake and Kubernetes is a plus.

• Engineering mindset: A focus on reliability, maintainability and continuous improvement, with an appetite to learn from more experienced colleagues and share what you learn.