Principal Applied Scientist
Core
Designing and training relevance models (LLM fine-tuning, LTR) and building evaluation pipelines to improve Copilot Chat, Copilot Search, and BizChat experiences.
Role type
Principal Applied Scientist (Search & Relevance)
Builds
Scalable search, retrieval, and ranking systems for enterprise Copilot products
Domain
Enterprise AI, Search, Information Retrieval, NLP
Deliverable
production ML models
Required skills
LLM fine-tuning, Learning-to-Rank (LTR), Retrieval-Augmented Generation (RAG), semantic search, information retrieval, A/B experimentation, data analysis at scale, research program leadership, technical strategy translation
Preferred skills
Hybrid search strategies, query rewriting, document reranking, conference speaking, patent/publication authorship
Technologies
OpenAI GPT, T5, Llama, LTR frameworks
Responsibilities
Lead end-to-end applied science projects from ideation to shipping; mentor and guide a team of applied scientists; invent and apply cutting-edge ML/NLP techniques for retrieval and ranking; translate business goals into scientific strategy; collaborate cross-functionally with engineering and product teams.
Seniority
Principal, strategy & mentorship
