Personalised Search & Retrieval Engineer (Agentic Commerce)

Swap; Greater London; On-site; GBP 55,000 - 85,000; Full time

Job summary

A leading software provider in London is seeking a Personalised Search & Retrieval Engineer to optimise AI search infrastructure. You will design embedding models and build personalisation systems, ensuring relevance and context in search results. Ideal candidates have over 4 years of experience in search engineering, strong knowledge of semantic search technologies, and proficiency in Python. This role offers a collaborative work environment committed to innovation and creativity.

Qualifications

• 4+ years of experience in search engineering.
• Deep expertise in semantic search technologies.
• Strong knowledge of search fundamentals and evaluation metrics.
• Hands-on experience with machine learning frameworks.
• Proficiency in Python and SQL.

Responsibilities

• Own Semantic Search Quality and optimise embedding models.
• Build personalisation systems to integrate user-level signals.
• Design reranking and relevance models.
• Create comprehensive evaluation frameworks.
• Optimise performance across the search pipeline.

Skills

Search engineering

Information retrieval

Semantic search technologies

Python

SQL

Machine learning frameworks

Distributed computing

Job description

Personalised Search & Retrieval Engineer (Agentic Commerce)

London

Company

Say hello to the ecommerce OS.

Swap is a leading software provider dedicated to empowering e-commerce brands with innovative, data‑driven solutions. Our cutting‑edge platform helps online retailers optimise their operations, enhance customer experiences, and drive growth. We are committed to fostering a collaborative and inclusive work environment where creativity and innovation thrive.

We are seeking a Personalised Search & Retrieval Engineer to own and optimise the core search infrastructure that powers personalised AI experiences. You'll be responsible for the entire search pipeline from semantic understanding to relevance ranking, building systems that deliver contextually relevant results whilst learning from user interactions to continuously improve search quality.

What You'll Do

• Own Semantic Search Quality: Design and optimise embedding models, chunking strategies, and indexing approaches to ensure high-quality semantic understanding and retrieval across diverse content types
• Build Personalisation Systems: Integrate user-level signals, behavioural patterns, and preference data into search algorithms to deliver contextually relevant and personalised results
• Design Reranking & Relevance Models: Combine BM25, dense retrieval, and learning‑to‑rank approaches. Build sophisticated reranking algorithms that incorporate relevance signals, user context, and business objectives to optimise search result ordering
• Create Evaluation Frameworks: Develop comprehensive evaluation systems including offline metrics, online A/B testing, and user satisfaction measurements to continuously assess and improve search performance
• Optimise Performance & Scale: Tune latency and throughput across the entire search pipeline, working with vector databases and distributed systems to ensure sub‑second response times
• Build Agentic Integrations: Design retrieval systems that provide accurate context for AI applications, ensuring responses are well‑grounded in relevant, up‑to‑date information
• Implement Continuous Learning: Create feedback loops that capture user interactions, click‑through rates, and engagement signals to continuously refine personalisation algorithms

Skills & Qualifications

• 4+ years of experience in search engineering, information retrieval, with proven track record of improving search quality at scale
• Deep expertise in semantic search technologies including embedding models, vector databases (postgres, pgvector), and modern retrieval architectures
• Strong knowledge of search fundamentals: BM25, TF‑IDF, learning‑to‑rank algorithms, and experience with search evaluation metrics (NDCG, MRR, MAP)
• Hands‑on experience with machine learning frameworks (PyTorch, TensorFlow), search libraries, and modern ML infrastructure
• Proficiency in Python, SQL, and distributed computing with experience building high‑throughput, low‑latency systems
• Experience with personalisation techniques, collaborative filtering, and integrating user behaviour signals into ranking algorithms
• Knowledge of A/B testing methodologies, statistical analysis, and search analytics for continuous optimisation
• Familiarity with modern AI/ML operations, model deployment, and monitoring in production environments