Senior ML Ops Engineer in London

This role is part of our Global:IQ team, the group developing our new intelligence platform. Global:IQ brings together a suite of 1st party and partner data, tools and capabilities to turn data into audience understanding and optimised, data‑led media plans. Using a combination of data science, machine learning & AI techniques, it supports smarter targeting across Global's audio and out‑of‑home inventory, optimises advertising creatives and automates the tracking of outcomes for advertisers through the acquisition funnel—from building awareness and consideration to driving action.

As a Senior MLOps Engineer you will play a critical role in building the operational infrastructure that brings AI and ML models into production at Global:IQ. You will own the platforms, pipelines and processes that enable Data Science and Applied ML teams to deploy, monitor, retrain and govern models reliably at scale across our ad‑targeting, creative optimisation and advertising measurement capabilities. This position demands a hands‑on engineer with deep expertise in operationalising machine learning systems, a strong understanding of cloud infrastructure, ML lifecycle management and production monitoring. You will work closely with Data Scientists, Data Engineers and Product teams to create robust, scalable and maintainable MLOps workflows—from the ground up. The role reports to the Lead Engineer (MLOps & AI) and is a unique opportunity to establish MLOps best practices in a truly innovative AI/ML & data‑driven product environment.

3 best things about the job

• Build from Zero: You're not maintaining legacy systems—you're establishing the MLOps patterns, tooling and standards that will scale with the team for years to come.
• AI at the Core: This is a true AI/Data‑driven product. ML isn’t a nice‑to‑have feature – it's the product. Your infrastructure directly enables business value.
• Truly Cross‑functional: The Global:IQ team is a tight collaboration between technical and commercial areas.

Measures of Success

• Defined a clear operating model between Data Engineering/MLOps and teams responsible for model development.
• Onboarded key 1st and 3rd party datasets following existing ingestion patterns/standards.
• Delivered an initial end‑to‑end MLOps path for at least one production ML use case, from model handoff through deployment, monitoring and rollback.
• Established baseline operational standards including model versioning, environment management, deployment patterns and handover processes between Data Science and Engineering.
• Implemented monitoring and alerting for production ML workloads, covering operational health, data quality and model performance signals.
• Defined a clear operating model and interfaces between teams developing models and teams operating them in production.
• Built collaborative relationships with Data Science, Data Engineering and Product stakeholders, demonstrating pragmatic judgement and delivery pace.

Key Responsibilities of the Role

• ML Infrastructure & Deployment (40%): Design, build and maintain automated pipelines for model training, validation, packaging and deployment across development, staging and production environments. Implement model registries, experiment tracking and versioning systems to ensure reproducibility and traceability. Build and operationalise batch, streaming and near‑real‑time inference services depending on product requirements. Create reusable patterns and self‑serve tooling that enable Data Science teams to deploy models independently while adhering to operational standards. Implement infrastructure for feature engineering pipelines, feature stores, and consistent serving layers between training and inference.
• Model Monitoring & Operations (30%): Implement comprehensive monitoring for ML workloads including prediction latency, throughput, error rates, input data quality and feature drift. Build alerting and automated recovery mechanisms to ensure SLAs for ML services are met. Establish processes for model rollback, rollout strategies (canary, blue‑green) and incident response. Define and track operational KPIs for ML systems and lead post‑incident reviews to drive continuous improvement. Work with Data Science teams to implement model performance tracking, drift detection and retraining triggers.
• MLOps Governance & Best Practice (20%): Establish governance controls for model lineage, approval workflows, reproducibility and audit trails. Define and enforce standards for model packaging, environment management, dependency management and code quality for ML workloads. Introduce CI/CD patterns specific to ML including automated testing (unit, integration, model validation), promotion gates and release automation. Document MLOps processes, runbooks and architectural decisions, enabling knowledge sharing across the team. Stay current with industry trends and tooling, evaluating and adopting fit‑for‑purpose technologies as the platform matures.
• Collaboration & Enablement (10%): Partner closely with Data Scientists and ML Engineers to understand requirements and translate experimental work into production‑ready systems. Work with Data Engineering teams to ensure data pipelines, governance and quality controls support ML use cases effectively. Support and mentor junior engineers, raising the bar on MLOps practices and operational excellence across the team. Communicate clearly with non‑technical stakeholders about ML system health, risks and trade‑offs.

What you will need

The ideal candidate will be pragmatic, hands‑on, and passionate about making ML systems reliable, scalable and maintainable in production.

Essential Skills & Experience

• Strong programming skills (Python preferred) with a focus on production‑quality, testable and maintainable code.
• Hands‑on MLOps experience: You have operationalised ML models in production, owning deployment, monitoring and lifecycle management (not just experimentation).
• Cloud platform expertise (AWS strongly preferred; Snowflake a plus) with deep understanding of services for compute, orchestration, storage and ML (e.g., SageMaker, Lambda, ECS/EKS, Step Functions).
• Experience with MLOps tooling such as: Experiment tracking and model registries (MLflow, Weights & Biases, SageMaker Model Registry), Workflow orchestration (Airflow, Prefect, Step Functions), Model serving frameworks (SageMaker, TorchServe, TensorFlow Serving, or similar), Feature stores (Feast, Tecton, or custom‑built).
• Deep understanding of monitoring and observability for ML systems, including operational metrics, data quality checks, drift detection and model performance tracking.
• CI/CD and Infrastructure as Code: Experience with ML‑specific CI/CD patterns, Terraform, containerisation (Docker), and testing automation for ML pipelines.
• Practical experience building MLOps from an early stage, including sensible tool selection, pattern definition and iterative delivery.
• Ability to work across disciplines: You can translate between Data Science language and Engineering standards, establishing clear contracts and interfaces.
• Strong communication skills: You can explain technical decisions, trade‑offs and system behaviour to both technical and non‑technical audiences.
• Analytical and data‑driven mindset: You use metrics, logs and evidence to diagnose issues and make decisions.

Desirable Skills & Experience

• Experience with agentic and AI‑accelerated coding tools (e.g., GitHub Copilot, Cursor, Claude Code) to increase delivery pace.
• Understanding of ML model types and performance characteristics (e.g., classification, regression, recommendation systems, forecasting, LLMs).
• Familiarity with advertising technology, marketing analytics or media measurement use cases.
• Experience in early‑stage or scale‑up environments where you've built foundational capabilities that grew with the team.
• Knowledge of data governance, privacy and compliance requirements in data‑driven products.

Personal Attributes

• Pragmatic builder: You balance speed with quality, making sensible trade‑offs and avoiding over‑engineering.
• Ownership mentality: You take responsibility for systems in production, including being on‑call and driving issues to resolution.
• Collaborative mindset: You work effectively across teams, valuing diverse perspectives and building trust through delivery.
• Curiosity and learning: You stay current with MLOps trends and aren't afraid to try new tools or approaches when they add value.
• Domain passion: You must love the challenge of using data & intelligence to drive ad campaign efficiency and demonstrate the value of the investment in the media. We are looking for someone curious about this domain, enthused to work on it for a number of years, and with the ambition to launch things that have never been done before.

Everyone is welcome at Global. Just like our media and entertainment platforms are for everyone, so are our workplaces. We know that we can't possibly serve our diverse audiences without first nurturing and celebrating it in our people and that's why we work hard to create an inclusive culture for everyone. We believe that difference will set us apart, so no matter what you look like, where you come from or what your favourite radio station is, we want to hear from you.