Back to Jobs
Search / Relevance Engineer (AI-Native Product)
Apply
Search / Relevance Engineer (AI-Native Product)
We’re partnering with a high-growth AI-native startup building a new category of revenue and GTM infrastructure for modern companies. They’re replacing fragmented legacy systems with a unified, intelligent platform used by fast-scaling startups.
They’re now looking for a Search / Relevance Engineer to own and scale the core systems that power retrieval, ranking, and semantic understanding across their product.
This is a foundational role focused on building search that is fast, context-aware, and deeply aligned with user intent across both structured and unstructured data.
What you’ll do
You’ll own and evolve the full search stack - from indexing to ranking to evaluation:
- Design and improve end-to-end search systems (indexing → retrieval → ranking → evaluation)
- Own search relevance and quality metrics (precision, recall, NDCG, and related IR metrics)
- Build and optimise indexing pipelines (tokenisation, normalisation, schema design, data modelling)
- Develop and iterate on ranking strategies:
- Keyword-based retrieval (BM25, hybrid search)
- Embedding-based retrieval (dense / vector search)
- Learning-to-rank and ML-based ranking models
- Work with modern embedding and semantic search systems (e.g. OpenAI embeddings, Sentence Transformers, vector databases)
- Design hybrid retrieval systems combining lexical + semantic approaches
- Build evaluation frameworks:
- Offline benchmarks and labelled datasets
- Human relevance judgement pipelines
- A/B testing and live experimentation
- Analyse search logs to identify failure modes, relevance gaps, and ranking issues
- Partner closely with product and data teams to align search quality with real user intent and product outcomes
What we’re looking for
- Strong experience building production search or relevance systems
- Deep understanding of information retrieval fundamentals (ranking, retrieval, evaluation)
- Experience with hybrid search systems and vector-based retrieval
- Strong backend and data systems capability, with exposure to ML-driven ranking approaches
- Analytical mindset with experience running experiments and iterating on relevance quality
Nice to have
- Experience from search / infra-heavy companies (e.g. Elastic, Redis, MongoDB, similar environments)
- Exposure to LLM-based retrieval or AI-native search systems
- Experience in early-stage or high-ownership environments