Senior Machine Learning Engineer: ML Recall

This position is listed on behalf of a partner company, who manages all applications and next steps. Our partner is looking for a Senior Machine Learning Engineer: ML Recall based in United Kingdom.

As part of the ML Recall team, you will help shape the retrieval layer of a large-scale e-commerce search engine used by hundreds of millions of shoppers worldwide.

You will build and improve retrieval systems that ensure relevant products are surfaced for queries across languages, markets, and domains.

Your work will span dense and sparse retrieval, query understanding, visual search, and multimodal solutions.

You will work with modern deep learning and transformer-based models, balancing relevance, generalization, and millisecond-level latency.

The role offers substantial room for research and experimentation, including testing new architectures and approaches on live traffic.

You will own models end to end, from problem definition and experimentation through deployment, A/B testing, and measurable impact.

You will join a highly technical, remote-first environment where your work directly influences search quality, user experience, and business performance.

Accountabilities

• Develop and optimize modern search retrieval systems using dense and sparse models, query understanding, and other machine learning techniques.
• Build solutions that balance retrieval quality and strict latency requirements, ensuring relevant products are returned within milliseconds.
• Develop end-to-end visual and multimodal search capabilities, including image search, visual recommendations, and "shop the look" experiences.
• Train, fine-tune, evaluate, deploy, and continuously improve deep learning models used across multiple products and teams.
• Experiment with new model architectures, retrieval approaches, and techniques, validating their effectiveness through A/B testing and live traffic experiments.
• Address complex measurement challenges in search recall by developing reliable ground truth and evaluation methodologies for products that may otherwise be missed entirely.
• Design models that generalize across 40+ languages and 20+ domains without relying on customer-specific rules or overrides.
• Take ownership of machine learning initiatives from problem framing and research through production rollout and ongoing optimization.
• Collaborate with other engineering and machine learning teams whose products build on the models and capabilities you develop.