Data Scientist - RecSys
ARRISE is seeking a Data Scientist to focus on recommendation systems. The role involves designing, implementing, and optimising end-to-end recommendation pipelines, ranging from initial data ingestion to final model inference.
The successful candidate will develop and refine machine learning models, research state-of-the-art approaches to enhance recommendation quality, and ensure the scalability of data pipelines. This position requires close collaboration with data and software engineering teams to deliver production-ready solutions that drive key business metrics.
- Design, implement, and optimise end-to-end recommendation pipelines, from data ingestion to model inference.
- Build and maintain scalable ETL pipelines to support reliable and efficient data flows.
- Develop, evaluate, and continuously improve ML models for recommendation systems.
- Research, prototype, and implement state-of-the-art approaches to improve recommendation quality and drive key business metrics.
- Scale and optimise data and model pipelines to handle large volumes of data and real-time or batch processing needs.
- Integrate multi-modal data (e.g., behavioural, transactional, and contextual signals) from various systems into recommendation models.
- Ensure robustness and stability of pipelines by implementing unit and integration tests across data, modelling, and deployment workflows.
- Monitor and maintain end-to-end system performance, including data pipelines, model quality, and downstream impact.
- Design and analyse A/B tests to evaluate model performance and support data-driven product decisions.
- Build dashboards and observability tools to track model metrics, system health, and business KPIs.
- Collaborate closely with Data Engineers, Software Engineers, and stakeholders to deliver scalable, production-ready solutions.
- Bachelor's or Master's degree in Computer Science, Engineering, or a related field (required)
- Strong Python experience with recent production use, including hands-on work with data science and machine learning libraries and frameworks (e.g., Pandas, Polars, NumPy, scikit-learn, PyTorch, TensorFlow, JAX, Hugging Face, â¦) (required)
- Experience building and deploying end-to-end machine learning systems on cloud AI platforms (Azure, GCP, or AWS), from ETL pipelines to deployment and monitoring, including model versioning and experiment tracking, supporting either batch or real-time workflows (required)
- Strong understanding of deep learningâbased recommender systems for next-item prediction, and analogous NLP architectures that model sequential patterns and context (required)
- Demonstrated experience building efficient data transformation pipelines for both transactional (OLTP) and analytical (OLAP) workloads, with strong knowledge of SQL and NoSQL databases (e.g., PostgreSQL, MySQL, Redshift, Snowflake, BigQuery, MongoDB, Cassandra) (required)
- Experience with unit and integration testing (e.g., Pytest), CI/CD pipelines, and Docker-based containerisation (required)
- Experience building large-scale recommender systems (e.g., candidate generation, ranking, retrieval, personalization) (nice-to-have)
- Track record of publications in deep learning at relevant conferences or journals (nice-to-have)
- Experience with Azure Data Factory / AWS Glue / Google Cloud Dataflow (nice-to-have)
- Experience designing and analyzing A/B tests, with a solid understanding of relevant evaluation metrics (nice-to-have)
- Experience designing and implementing metadata-driven pipelines to scale automated A/B testing systems (nice-to-have)
- Experience developing multi-modal models that integrate multiple data types (e.g., text, images, audio) (nice-to-have)
- Experience applying transformer-based models or large language models (LLMs) to recommendation or personalization tasks (nice-to-have)
- Experience with distributed training, including data parallelism and model parallelism (nice-to-have)
- Experience with distributed data processing and big data technologies (e.g., Spark, Hadoop, Flink, Kafka, Hive, Presto, Databricks) (nice-to-have)
Pragmatic Play is a leading content provider to the iGaming industry, delivering a multi-product portfolio of online slots, live casino, bingo and virtual sports to operators worldwide. Founded in 2015 and headquartered in Gibraltar, it releases new titles at a high cadence and is certified across numerous regulated markets. It operates as part of the ARRISE group, which runs the live-casino studios and operational hubs across Europe, the Middle East and Latin America that power its games.