Senior Data Scientist I – LeapSpace

About the role

We are seeking a Senior Data Scientist I to lead the development and evaluation of advanced search and generative AI systems within LeapSpace. The role involves end‑to‑end ownership of complex problem areas, driving methodological rigor, and shaping the technical direction of retrieval and RAG solutions.

Key responsibilities

• Design and optimise lexical, vector and hybrid retrieval systems at scale.
• Architect and improve Retrieval‑Augmented Generation (RAG) pipelines, including retrieval strategies, prompt design and system orchestration (e.g., LangGraph‑based workflows).
• Experiment with embeddings, re‑ranking models and retrieval architectures to boost relevance and user outcomes.
• Partner with engineering teams to ensure robust, scalable, production‑ready implementations.
• Define and evolve evaluation strategies for search and generative AI across products.
• Develop robust IR evaluation frameworks (NDCG, recall, ranking quality) and GenAI evaluation metrics (grounding, faithfulness, hallucination detection).
• Create evaluation datasets, gold standards and annotation pipelines.
• Guide experimental design, including offline evaluation and A/B testing, ensuring statistical rigour.
• Contribute to responsible AI practices, addressing bias, fairness and risk.
• Apply state‑of‑the‑art NLP, embeddings and generative AI techniques to production use cases.
• Integrate emerging technologies into the team roadmap.
• Support knowledge‑graph and semantic enrichment efforts that enhance retrieval systems.
• Collaborate with domain experts, ontology engineers and biomedical informaticians to incorporate scientific taxonomies and citation networks.

Required profile

• Deep hands‑on experience with search/retrieval systems and RAG pipelines.
• Proven track record designing and deploying production‑grade AI solutions.
• Strong expertise in evaluation frameworks and statistical experiment design.
• Commitment to responsible AI, including bias and fairness assessment.
• Ability to work independently as a senior individual contributor while influencing technical direction.

Required skills

• Lexical, vector and hybrid retrieval techniques
• Retrieval‑Augmented Generation (RAG) pipeline development
• LangGraph workflow orchestration
• Embeddings and re‑ranking model implementation
• IR evaluation metrics (NDCG, recall, ranking quality)
• Generative AI evaluation (grounding, faithfulness, hallucination detection)
• Knowledge‑graph construction and semantic enrichment
• Prompt design for large language models
• Statistical experiment design and A/B testing