ML engineer in recommender systems

Aristocrat; Barcelona; Hybride; EUR 70 000 - 95 000; Plein temps

Résumé du poste

Aristocrat ищет ML Engineer для команды MLOps, чтобы усилить моделировочное ядро и обеспечить устойчивость моделей в продуктивной среде. Роль фокусируется на рекомендательных системах и производительности, взаимодействии с командами продукта и данных.

Требуется 4+ лет опыта, владение Python, ML-библиотеками и опытом работы в облачных платформах. Гибридный формат работы, минимум 3 дня в офисе в неделю.

Qualifications

• 4+ лет применения ML к реальным задачам, от данных до развёртывания.
• Опыт в рекомендательных системах (collaborative filtering, ranking) на реальных продуктах.
• Сильные навыки Python и библиотеки ML (scikit-learn, PyTorch/TensorFlow, XGBoost).
• SQL-опыт и работа с большими дата- warehouse.
• Опыт трекинга экспериментов и/или реестра моделей (MLflow, Weights & Biases).
• Навыки мониторинга моделей в продакшене и работы с облаками (GCP/AWS/Azure).
• Опыт работы с Docker и Airflow.
• Сильная аналитическая база в математике/инженерии/CS.

Responsabilités

• Разрабатывать, обучать, оценивать и пересматривать ML-модели с фокусом на рекомендатели.
• Определять метрики производительности, пороги и понятие «модель в рабочем состоянии» в бизнес-терминах.
• Содействовать экспериментации: версионирование моделей и отслеживание экспериментов (MLflow/Weights & Biases).
• Участвовать в формировании культуры governance: как продвигать модели в продакшн и коммуникационный контракт между MLOps и DS.
• Создавать шаблоны и каркасы структурирования Data Science проектов и моделей.
• Строить конвейеры обучения и подготовки признаков, писать чистый Python/SQL.
• Исследовать ранние проекты с LLMs, RAG и агентными AI как дополнение к роли.
• Разрабатывать простые интерфейсы (например, Streamlit) для пользователей без технических навыков.
• Сотрудничать с командами: продукт, данные и инженеры, ясно сообщать результаты и выводы.

Connaissances

Python

SQL

ML frameworks

Cloud platforms

Team collaboration

Outils

MLflow

Weights & Biases

Docker

Airflow

Streamlit

Description du poste

Описание:

At Aristocrat, we are advancing gaming technology with modern, powerful solutions. Our mission is to bring happiness to life through play. We value collaboration, creativity, and a strong passion for excellence. These roles primarily support the Machine Learning area of our Product Madness business line — a tech company specialised in social casino mobile games. We are looking for an ML Engineer to join our MLOps team and strengthen its modeling core.

Задачи:

• Design, train, evaluate and retrain machine learning models with emphasis on recommender systems — that support game features, player experience, and operational efficiency.


• Own the design of our model performance monitoring: define what to track (drift, data quality, degradation), thresholds, and what “the model is still healthy” means in business terms.


• Help establish a solid experimentation culture: model versioning and experiment tracking with tools such as MLflow or Weights & Biases, so experiments are reproducible and model versions are easy to compare and promote.


• Help establish a model governance culture: define how models are promoted to production and lay the foundations of the communication contract between the MLOps team and Data Science.


• Create templates and cookiecutter scaffolding to standardise how Data Science structures, hands off, and communicates its models and experiments.


• Build robust training and feature-engineering pipelines, and write clean, production-ready Python and SQL.


• Explore and contribute to early-stage projects involving LLMs, RAG and agentic AI where they bring value — as a complement to the role, not its core.


• Develop simple interfaces (e.g., Streamlit) to expose ML capabilities to non-technical users when useful.


• Collaborate with teams across product, data, and engineering, and communicate results, limitations, and recommendations clearly across technical and business audiences.

Требования:

• 4+ years of experience applying machine learning to real-world problems, from data to deployment.


• Proven experience with recommender systems (collaborative filtering, ranking, or similar) applied to real products.


• Strong Python skills and hands-on experience with ML libraries (scikit-learn, PyTorch/TensorFlow, XGBoost or similar).


• Solid SQL expertise and experience working with large data warehouses.


• Hands-on experience with experiment tracking and/or model registry tools (MLflow, Weights & Biases, or equivalent).


• Experience designing or contributing to model monitoring / performance tracking in production.


• Comfort working with cloud platforms (e.g., GCP, AWS, Azure), Docker, and Airflow.


• A strong analytical foundation, with a background in a quantitative field (mathematics, physics, computer science, engineering) or equivalent experience.


• Будет плюсом: Experience with Snowflake or similar cloud data warehouses;


• Будет плюсом: Hands-on experience building applications with Generative AI (LLMs), including API integration;


• Будет плюсом: Interest or early experience in RAG systems — whether through prototypes, side projects, or professional work;


• Будет плюсом: Interest or familiarity with agentic AI concepts or frameworks (e.g., LangChain agents, CrewAI);


• Будет плюсом: Ability to build simple user interfaces (e.g., Streamlit) to expose ML models to wider teams

Условия:

• Minimal travel required (up to 5%)


• Hybrid role with 3 days at the office weekly.

We offer a comprehensive pay and benefits package designed to stay competitive in the market, support your wellbeing, and recognise your contribution to our success. Your specific compensation package will be determined by factors such as your skills, experience, qualifications, and location. Depending on your role and location, you may be eligible for annual bonuses and incentives, health and wellbeing benefits, paid time off, retirement plans, insurance coverage, and other local or statutory benefits.