Data Scientist (Search & Recommendations) at Mayflower

on-site · full_time · Visa sponsorship

Apply for this role at Mayflower

Responsibilities:
- Design, develop and improve large-scale retrieval pipelines (candidate generation, query processing, matching, filtering).
- Build and optimise ranking models and learning-to-rank solutions for low-latency production inference.
- Develop recommendation pipelines (recall + ranking) for personalised and non-personalised use cases.
- Extract and engineer user/item/session/context features; create reproducible feature and model pipelines.
- Design and run offline and online experiments (A/B tests); evaluate using CTR, conversion, NDCG, MRR and related metrics.
- Collaborate with product, backend, data engineering and MLOps teams to deploy, monitor and maintain models.

Requirements:
- Proven experience building production search, ranking or recommendation systems at scale.
- Strong Python and SQL skills for data processing, model training and analysis.
- Familiarity with ML frameworks (PyTorch/TensorFlow/XGBoost) and evaluation metrics for relevance and ranking.
- Experience with high-load, real-time systems and streaming platforms (e.g., Kafka) and distributed data processing.
- Practical knowledge of experiment design, model validation, monitoring and addressing bias/robustness issues.
- Good communication skills and ability to collaborate across engineering and product teams.