ML Engineer

Description

This role sits in the ML group of the social casino business unit, a free-to-play mobile game publisher within a company of over 7,300 employees worldwide across three business units.

You will join the MLOps team to strengthen its modelling core, owning models from data and experiments through to production.

You will also help keep models reliable and aligned with business goals over time by building up a culture of monitoring and experimentation.

Your work will shape game features and the experience players have.

Responsibilities

• Build, train, evaluate and retrain models, mainly recommender systems, that power game features, player experience and operational efficiency.
• Design performance monitoring for models: choose what to track, such as drift, data quality and degradation, set thresholds, and define what a healthy model means in business terms.
• Help set up experiment tracking and model versioning with MLflow or Weights & Biases, so experiments can be reproduced and versions compared and promoted easily.
• Help define governance, including how models move into production and the agreed interface between MLOps and Data Science.
• Build templates and cookiecutter scaffolds so Data Science structures, hands over and reports its models and experiments in a common way.
• Create robust training and feature pipelines, writing clean, production-ready Python and SQL.
• Contribute to early-stage LLM, RAG and agentic AI projects where they add value, as a side focus rather than the core of the role.
• Build simple Streamlit interfaces that give non-technical users access to ML capabilities when useful.
• Work with product, data and engineering teams, explaining results, limitations and recommendations to both technical and business audiences.

Requirements

• 4+ years applying machine learning to real problems, from data through to deployment.
• Proven track record with recommender systems, such as collaborative filtering or ranking, in live products.
• Strong Python and hands-on use of ML libraries such as scikit-learn, PyTorch or TensorFlow, and XGBoost or similar.
• Solid SQL and experience working with large data warehouses.
• Hands-on work with experiment tracking or model registry tools such as MLflow, Weights & Biases or equivalent.
• Experience designing or contributing to monitoring of models in production.
• Comfortable with a cloud platform such as GCP, AWS or Azure, plus Docker and Airflow.
• Strong analytical grounding in a quantitative field such as mathematics, physics, computer science or engineering, or equivalent experience.
• Experience with Snowflake or a similar cloud data warehouse.
• Hands-on experience building applications with generative AI (LLMs), including API integration.
• Interest in or early experience with RAG systems, through prototypes, side projects or professional work.
• Interest in or familiarity with agentic AI concepts or frameworks such as LangChain agents or CrewAI.
• Ability to build simple user interfaces, such as Streamlit, to expose ML models to wider teams.

Conditions

Hybrid role with 3 days a week in the office.

Travel is minimal, up to 5%.

Depending on the role, you may need to register with the Nevada Gaming Control Board (NGCB) and/or other gaming jurisdictions where the company operates.

Pay is set by your skills, experience, qualifications and location, and specific compensation details will be shared during recruitment.

The company rewards individual impact through a pay-for-performance approach.

Benefits

• You may be eligible for annual bonuses and incentives, depending on your role and location.
• Depending on your role and location, you may also get health and wellbeing benefits, paid time off, retirement plans, insurance coverage, and other local or statutory benefits.