We use PostHog and Clarity to improve the site, and Meta Pixel to measure page views and ad effectiveness. Usage data may include page addresses and browser identifiers. You can opt out of collection at any time. Details in the Privacy Policy
Senior Machine Learning Engineer at Tripledot Studios · Trevo
Who are we?
Tripledot Studios is one of the largest independent mobile games companies in the world.
We are a multi-award-winning organisation, with a global 2,500+ strong team across 12 studios.
Our expanded portfolio includes some of the biggest titles in mobile gaming, collectively reaching top chart positions around the world and engaging over 25 million daily active users.
Tripledot’s guiding principle is that when people love what they do, what they do will be loved by others.
We’re building a company we’re proud of. One filled with driven, incredibly smart and detail-orientated people, who LOVE making games.
Our ambition is to be the most successful games company in the world, and we’re just getting started.
Role Overview
Join Tripledot Studios as a Senior Machine Learning Engineer working on dynamic pricing and recommender systems use cases in our games. The work involves tabular machine learning, including neural networks and approaches such as gradient-boosted decision trees. You'll focus on understanding the data, developing features and improving the models, rather than primarily owning production deployment.
You'll work toward defined near-term ML targetds, collaborating with the monetization team and game product teams to plan A/B tests. Over the longer term, you'll work with those teams to translate product and monetization requirements into ML objectives.
You'll start by learning the project and its business goals. Within the first three months, we'd expect you to contribute to the training pipeline and investigate features and model issues. By around six months, the aim is for you to take a more proactive role in setting the model's direction.
What you'll do
Build and improve training pipelines for dynamic pricing and recommender system models, from feature and label design through training, tuning and offline evaluation of tabular models such as neural networks and gradient-boosted trees.
Monitor model performance once models are live across games and products. Investigate data and concept drift, including shifts tied to new titles, client versions or user behaviour, and make models easier to extend to another product.
Diagnose missed model outcomes across the training pipeline, business logic and underlying data, then work through the improvements needed to restore model quality.
Work with monetization and product colleagues to connect model decisions to revenue and player outcomes, and help plan the A/B tests that inform what ships.
As you get up to speed, identify gaps in the team's understanding of the models and propose improvements to their direction.
Requirements
Hands-on training and evaluation of tabular models, using neural networks or gradient-boosted trees. Experience with PyTorch, PyTorch Lightning, TensorFlow, XGBoost, CatBoost or scikit-learn could all be relevant; no single framework is required.
Experience designing features and labels, choosing metrics for the decision a model makes and judging when offline results warrant an A/B test.
Proficiency in SQL and the ability to write efficient queries to extract, manipulate and aggregate data from relational databases.
An investigative approach to incomplete data and longer-term requirements that need clarification.
The ability to explain model results to monetization and product partners in terms of revenue and player outcomes.
Experience in ad tech, recommender systems or online marketplaces would be useful, as would experience with production APIs or deploying ML models. Experience with the Ray framework would also be a plus. None of these is required for the role.
Uses AI-assisted development tools, including code assistants and LLM-based copilots, to accelerate implementation, debugging and iteration of machine learning systems while maintaining production-quality standards.
Critically reviews and validates AI-generated code, model implementations and infrastructure configurations for reliability, correctness and maintainability, and explores AI-powered approaches to improve developer productivity or ML platform capability.
Working for Tripledot
25 days paid holiday in addition to bank holidays to relax and refresh throughout the year
Hybrid Working
20 days remote working: Work from anywhere in the world, or use the time to cover mandatory office days to WFH, 20 days of the year.
Daily Free Lunch: when in the office you get £12 every day to order from JustEat.
Regular company events and rewards: quarterly on-site and off-site events that celebrate cultural events, our achievements and our team spirit.
Employee Assistance Program: Anytime you need it, tap into confidential, caring support with our Employee Assistance Program, always here to lend an ear and a helping hand.
Family Forming Support: Receive vital support on your family forming/ fertility journey with our support program [subject to policy]
Life Assurance & Group Income Cover: Financial protection for you and your loved ones.
Continuous Professional Development: Propel your career with continuous opportunities for professional development.
Private Medical Cover & Health Cash Plan: Opt-in (P11d benefit) comprehensive private medical cover with Bupa and cash plan with Medicash.
Dental Cover: Opt-in (P11d benefit).
Cycle to Work Scheme: Salary sacrifice bike purchase scheme.
Pension Plan: Qualifying earnings or opt-in 4% contributory match schemes offered.