Senior Machine Learning Engineer

Description

The company is one of the largest independent mobile games makers in the world.

It has more than 2,500 staff across 12 studios and has won multiple industry awards.

Its games include some of the biggest mobile titles, many reaching top chart positions worldwide and drawing over 25 million daily active users.

This is a senior machine learning role on dynamic pricing and recommender systems in the games. The work centres on tabular models, including neural networks and gradient-boosted decision trees.

The focus is on understanding the data, building features and improving models, rather than mainly running production deployment.

Near-term, you work toward set ML targets with the monetization team and game product teams, who also help plan A/B tests. Over time, you help turn product and monetization requirements into ML objectives.

In your first three months, you learn the project and its business goals, then contribute to the training pipeline and look into features and model issues. By about six months, you are expected to take a more active part in setting the model's direction.

Responsibilities

• Build and refine training pipelines for pricing and recommender models, covering feature and label design, training, tuning and offline evaluation of tabular models.
• After launch, watch model performance across games and products. Track data and concept drift, such as changes tied to new titles, client versions or player behaviour, and make models easier to adapt to another product.
• When models miss their targets, find the cause in the pipeline, the business logic or the underlying data, and make the fixes needed to restore quality.
• Partner with monetization and product colleagues to link model choices to revenue and player results, and help plan the A/B tests that decide what ships.
• Once settled in, find gaps in the team's understanding of the models and propose changes to their direction.

Requirements

• Hands-on experience training and evaluating tabular models with neural networks or gradient-boosted trees. Frameworks such as PyTorch, PyTorch Lightning, TensorFlow, XGBoost, CatBoost or scikit-learn may be relevant, but no single one is required.
• Experience designing features and labels, choosing metrics for a model's decision, and judging when offline results justify an A/B test.
• Strong SQL, with the ability to write efficient queries that extract, reshape and aggregate data from relational databases.
• A methodical approach to incomplete data and to requirements that still need clarifying.
• The ability to explain model results to monetization and product partners in terms of revenue and player outcomes.
• Useful, not required: background in ad tech, recommender systems or online marketplaces, and experience with production APIs or deploying ML models.
• A plus, not required: experience with the Ray framework.
• Use AI-assisted development tools, such as code assistants and LLM-based copilots, to speed up implementing, debugging and iterating on ML systems while keeping production-quality standards.
• Review and validate AI-generated code, model implementations and infrastructure configurations for reliability, correctness and maintainability. Explore AI-driven ways to improve developer productivity or ML platform capability.

Conditions

Hybrid working.

You get 20 days a year of remote working. Use them to work from anywhere in the world, or to cover mandatory office days by working from home.

Benefits

• 25 days paid holiday, plus bank holidays.
• Regular company events and rewards marking cultural occasions, company achievements and team spirit.
• Private medical cover.
• Life and critical illness cover.
• Family forming and fertility support through a programme, subject to policy.
• Confidential support through the Employee Assistance Program, available at any time.
• Sport compensation benefit.
• Meal and transport vouchers.
• English and Spanish language classes.
• Continuing professional development opportunities.