Applied Scientist 2

Share job

The M365 Copilot RAG team is building the next generation of intelligent assistant systems to empower enterprise users with advanced productivity capabilities. We are hiring multiple Applied Scientists to join our efforts in enterprise-grade search, agentic RAG and multi-agent orchestration. This role will directly contribute to the core capabilities of Copilot and impact how millions of users interact with their enterprise data.
Our team is composed of world-class scientists and engineers who are passionate about turning cutting-edge research into high-quality, scalable products and services. We welcome candidates at levels 61-62.

• Bachelor’s, Master’s, or PhD degree in Computer Science, Electrical Engineering, or related fields.
• Strong algorithmic and modeling skills, with solid problem-solving capabilities.
• Experience in NLP, RL, search, or recommendation systems is preferred.
• Familiarity with complex document understanding, RAG frameworks, and multi-agent orchestration is a strong plus.
• Excellent communication and collaboration skills, with the ability to thrive in cross-functional environments.

• Develop sub-agents for search, find, and related tasks over complex enterprise private data with diverse types and heavy schema.
• Optimize core algorithms for RAG systems, including retrieval, ranking, and grounding extraction.
• Explore novel task decomposition and execution strategies within multi-agent collaboration frameworks.
• Analyze offline and online performance signals to identify optimization opportunities and improve user experience and system efficiency.
• Collaborate closely with product and platform teams to deliver robust, scalable solutions.