About the Role
ATG has a wealth of data, with millions of guests visiting, eating, and drinking in our theatres every year, and there is a growing need to bring machine learning and data-driven decision-making into our core business processes. The focus of this role is to improve revenue, customer satisfaction, retention, and acquisition by delivering production-ready machine learning solutions and actionable insights from our customer data. You will join a small, growing ML team at a pivotal time for ATG, taking ownership of defined parts of our ML estate, and working towards wider ownership as you grow.
Responsibilities
•Take ownership of defined components of the revenue management application (i.e. regional pipelines, seat map and pricing data pulls, or model inference), including their reliability and ongoing improvement.
•Productionise and automate machine learning models end-to-end, including data ingestion, feature engineering, model training and evaluation, inference, and delivery of pricing recommendations.
•Build and maintain data pipelines and MLOps workflows (CI/CD, model versioning, reproducibility, runtime handling) in AWS and Snowflake.
•Support the diagnosis and resolution of production issues, including on-call style triage of failed pipeline runs and data quality problems, escalating to the Lead ML Engineer where needed.
•Design and implement monitoring for model outputs (data drift, prediction drift, anomaly detection) and build dashboards to track model behaviour and recommendation quality daily.
•Implement guardrails for pricing recommendations (thresholds, overrides, fallback logic) in line with agreed designs.
•Write well-tested, documented, maintainable Python and SQL, following team conventions and taking part in peer code review.
•Work with Data Science to take experimental models and prototypes into production, and with commercial teams to translate business needs into technical requirements.
•Document systems, runbooks, and model behaviour so that knowledge is shared across the team.
•Contribute to the design and build of LLM-powered solutions (e.g. internal tools, decision-support systems, automation workflows), including prompt management, versioning, and evaluation.
•Use our data platform to support and automate decision-making across the business, bringing data together to constantly improve our understanding of our products and customers.
•Be an active part of our data community as we build new capability across disciplines, sharing learning and contributing to team standards.
Requirements
•Proven experience as a machine learning engineer, data scientist, or similar role, with a track record of delivering ML work into production.
•Experience of building and optimising data science workloads in cloud environments (AWS preferred).
•Strong Python and SQL, including writing complex queries and well-structured, tested code.
•Experience deploying and maintaining ML models in production, with a working understanding of MLOps (CI/CD, versioning, reproducibility).
•Experience building monitoring, alerting, and observability for data or ML systems.
•Understanding of system reliability concepts (failures, retries, idempotency, fault tolerance) and an interest in designing safeguards for automated decision systems.
•Able to work effectively with data scientists and data engineers, and to communicate clearly with technical and non-technical colleagues and stakeholders.
•Able to manage your own workload on defined projects, ask for help early, and take on feedback and coaching.
Nice to Have
•Prior experience of a similar industry (limited capacity and variable demand).
•Experience with pricing, revenue optimisation, or demand forecasting.
•Experience building monitoring dashboards (e.g. Streamlit, BI tools).
•Experience working with LLMs or generative AI systems in a production or applied setting.
•Experience with Snowflake, Terraform or SageMaker pipelines.
•Interest in growing towards technical leadership, including mentoring, and owning a wider area of the platform.
Conditions
No conditions stated in the posting.
