Machine Learning Engineer

Machine Learning Engineer

~2 days a week in the London office (hybrid, there is a little flexibility on this)

About the Company

Join a fast-growing platform business that connects buyers and sellers across a diverse marketplace. With deep investments in data and personalisation, the company is building advanced machine learning capabilities to improve discovery, user experience, and recommendation quality at scale.

The environment is fast-moving, collaborative, and focused on bringing practical AI into live systems that support millions of transactions across unique and varied inventory types. You’ll work at the intersection of engineering, product, and data to help design recommendation infrastructure that learns and adapts in real time.

What You’ll Be Doing

• Develop recommendation systems using a blend of collaborative, content-based, and innovative techniques (reinforcement learning, LLMs, etc.) to help surface relevant results for users
• Work on learning-to-rank models that fine-tune search and browse experiences based on behavioural and contextual signals
• Create adaptive personalisation logic based on browsing behaviour, previous interactions, and user profiles
• Build item classification models, embeddings, and semantic search tools to enhance product discovery
• Integrate ML services into production with support from MLOps and platform engineering teams
• Run structured experiments including A/B tests and offline simulations to measure impact and identify optimisation opportunities
• Apply visual similarity and image-based features to improve ranking and content understanding where appropriate
• Stay current with best practices and research in recommendation science, model evaluation, and user engagement

What They’re Looking For

• Proven experience developing and deploying recommendation algorithms in a commercial setting
• Strong Python programming ability and comfort using ML libraries like PyTorch, TensorFlow, scikit-learn, and related tooling
• Understanding of ranking algorithms, embeddings, and modern architectures such as two-tower models or transformers
• Experience working with behavioural modelling, personalisation strategies, and contextual learning
• Familiarity with metrics used to evaluate recommender systems
• Good understanding of cold-start challenges, sparse data, and interaction-based feedback loops
• Exposure to vector search technologies and embedding-based infrastructure is desirable
• Comfort running experiments in production environments and interpreting results across key business and user goals
• Bonus: experience with e-commerce or multi-category platforms where discovery and relevance are central to the product

If this role interests you and you would like to find out more (or find out about other roles), please apply here or contact us via (feel free to include a CV for review).