Data Scientist | Permanent | Hybrid working in London | Beverage Trading Platform

The Role

You will drive the development of machine learning systems that power a global exchange and data platform. You will work at the intersection of NLP, recommendation systems, and time-series forecasting, building production-grade solutions that directly impact trading decisions. This role offers the opportunity to work on semantic search, hybrid recommendation engines, and predictive models.

What You’ll Do

• NLP & Search: Design and deploy models for entity matching, semantic search, and text classification.
• Recommender Systems: Build engines combining collaborative filtering, content-based filtering, and business rule layers.
• Forecasting: Develop time-series models to predict market trends and pricing dynamics.
• Transformer Models: Fine-tune and deploy models such as BERT and sentence transformers for production use.
• ML Pipelines: Implement pipelines on cloud infrastructure using PySpark for large-scale data processing.
• Collaboration: Work with engineers to integrate models via REST APIs and batch processing.

What We’re Looking For

• Experience: 3+ years in data science or ML engineering, taking projects from research to production.
• Education: Bachelor's degree or higher in a quantitative field (e.g., Computer Science, Mathematics, Statistics, Physics; PhD welcome).
• NLP Expertise: Hands-on experience with modern NLP, including transformer models, embeddings, and semantic search (RAG systems highly desirable).
• Foundational ML: Strong foundation in statistical learning, classical ML (random forests, gradient boosting), and model validation.
• Technical Skills: Excellent Python and SQL skills with a focus on writing scalable, production-ready code.
• Tools: Experience with cloud services (ideally AWS), Git, and software engineering practices like testing and documentation