Senior Applied Scientist

Microsoft•Redmond, WA

About The Position

The Senior Applied Scientist will advance search and retrieval quality for Microsoft 365 Copilot Chat, Copilot Search, Work IQ, and agent experiences. The role focuses on modernizing the search stack for agentic and traditional search in enterprise environments by applying large language models and improving sparse, dense, and multimodal retrieval, query understanding, and related machine-learning methods. The scientist will collaborate across research, engineering, platform, and product teams to adapt promising advances to product needs and deploy effective approaches at scale. The position owns research-backed solutions from problem formulation through deployment. It involves defining and measuring retrieval quality through representative datasets, rigorous offline evaluation, statistical analysis, and production experiments; balancing relevance, latency, and efficiency; monitoring deployed models; and iterating as results or data change. The role also provides scientific leadership, shares findings, mentors less experienced scientists, and may contribute to publications or intellectual property. Microsoft describes a collaborative, fast-paced environment that brings together expertise in large language models, deep learning, information retrieval, and software engineering. The company emphasizes a growth mindset, collaboration, respect, integrity, accountability, and inclusion.

Requirements

Bachelor's degree in Statistics, Econometrics, Computer Science, Electrical or Computer Engineering, or a related field and 4+ years of related experience; or a master's degree in one of these fields and 3+ years of related experience; or a doctorate in one of these fields and 1+ year of related experience; or equivalent experience.
Ability to meet Microsoft, customer, and/or government security screening requirements, including passing the Microsoft Cloud Background Check upon hire or transfer and every two years thereafter.
Experience applying machine learning to information retrieval, search relevance, or related retrieval problems.
Experience shipping and evaluating deep-learning models in production at scale, including diagnosing quality or performance regressions.
Solid programming skills in Python and experience with a modern deep-learning framework such as PyTorch.
Experience designing offline evaluations and interpreting online experiments to make product decisions.
Ability to lead an ambiguous technical problem and collaborate with engineering and product teams to deliver a solution.

Nice To Haves

Master's degree in Statistics, Econometrics, Computer Science, Electrical or Computer Engineering, or a related field and 6+ years of related experience; or doctorate in one of these fields and 3+ years of related experience; or equivalent experience.
3+ years of experience creating publications, such as patents, libraries, or peer-reviewed academic papers.
Experience presenting at conferences or other events in the outside research or industry community as an invited speaker.
3+ years of experience conducting research as part of an academic or industry research program.
1+ year of experience developing and deploying live production systems as part of a product team.
1+ year of experience developing and deploying products or systems at multiple points in the product cycle, from ideation to shipping.
Experience with agentic search, including query planning, iterative retrieval, evidence selection, and evaluating the quality and efficiency of grounded responses.

Responsibilities

Lead applied research to improve retrieval quality for Microsoft 365 Copilot Chat, Copilot Search, Work IQ, and agent experiences.
Evaluate advances in information retrieval, natural language processing, and large language models, and adapt promising approaches to enterprise data and product needs.
Modernize the search stack for agentic and traditional search.
Develop and improve sparse and dense retrieval methods, enterprise-adapted models, and query-understanding techniques.
Partner with engineering and platform teams to put effective approaches into production at scale.
Define and measure retrieval quality using representative datasets, offline evaluation, statistical analysis, and production experiments.
Identify quality gaps, test hypotheses, understand model behavior, and communicate results that guide product decisions.
Own research-backed solutions from problem formulation through deployment.
Balance relevance, latency, and efficiency; monitor deployed models against quality and performance expectations; and iterate when results or data change.
Provide scientific leadership across disciplines by partnering with product, engineering, and research teams, shaping priorities, sharing practical advances, and mentoring less experienced scientists.
Document methods and findings, and contribute to publications or intellectual property where appropriate.