Join a pioneering team at Microsoft that's shaping the future of enterprise knowledge integration. We are building systems to transform data from critical applications like ServiceNow, Salesforce, and GitHub into intelligent knowledge for Microsoft 365 Copilot and other AI experiences. This role offers a unique chance to define the scientific underpinnings of retrieval, grounding, and agent-ready knowledge systems that will power the next wave of AI innovation.

Our work spans the entire stack, from data ingestion and modeling to indexing, querying, and integration. We manage a vast ecosystem of connectors and build sophisticated agentic tools to ensure scalability and efficiency. As a Principal Applied Scientist, you will be instrumental in defining the scientific direction for these advanced systems.

Drive cutting-edge research in enterprise retrieval, ranking, grounding, and knowledge representation. Develop novel techniques for autonomous workflows that empower AI agents and next-generation Microsoft AI experiences.

Advance retrieval quality through robust evaluation frameworks and metrics, focusing on recall, grounding accuracy, citation correctness, freshness, coverage, and task success. Explore advanced concepts like multi-hop retrieval, cross-source reasoning, knowledge graphs, and agent memory systems.

Build trust-aware AI systems that rigorously adhere to enterprise permissions and authorization protocols. Pioneer autonomous knowledge acquisition processes, including relationship discovery and metadata inference.

Collaborate closely with engineering and product teams to translate groundbreaking scientific discoveries into production-ready, hyperscale solutions. You will influence platform architecture and align roadmaps across multiple teams.

We seek candidates with a strong academic background and significant practical experience. A Bachelor's Degree in Statistics, Econometrics, Computer Science, Electrical or Computer Engineering, or a related field, coupled with 6+ years of relevant experience in areas like statistics or predictive analytics, is a minimum requirement. Alternatively, a Master's Degree with 4+ years of experience or a Doctorate with 3+ years of experience, or equivalent practical experience, will be considered.

A proven ability to translate complex research concepts into large-scale production systems is essential. Preferred qualifications include a Master's Degree with 6+ years of experience or a Doctorate with 5+ years of experience. Expertise in enterprise AI, retrieval systems, agent architectures, knowledge platforms, and building comprehensive evaluation frameworks for agentic systems is highly valued.

Additional requirements include the ability to meet Microsoft's security screening standards, including a Microsoft Cloud Background Check. The role requires 8+ years of experience across Machine Learning, demonstrating a strong track record of applying ML in production environments. Experience in the enterprise space and the proven ability to influence platform architecture and align cross-team roadmaps are also key.