New 
Research Engineer
 Microsoft | |
 United States, Washington, Redmond | |
 Oct 31, 2025 | |
|
OverviewJoin us in building a cutting-edge platform that enables seamless observation and interaction with AI agents. Our mission is to empower creators and organizations by providing tools for Human-in-the-Loop (HITL) workflows, intelligent prioritization, and advanced learning systems. You'll be part of a team that thrives on innovation and collaboration to redefine how humans and AI work together. As a Research Engineer, you will design and implement research-driven solutions that enhance agent transparency, task grouping, and adaptive learning. You'll explore methods for integrating memory, knowledge graphs, evaluations, and AI judges into our platform to improve agent performance and user experience. This role offers opportunities to deepen your expertise in applied AI, contribute to scalable systems, and shape the future of human-AI collaboration. You will work closely with internal and external stakeholders to develop algorithms that analyze interactions and generate actionable insights. Your contributions will help create teaching and learning moments for agent makers, driving continuous improvement. This position provides a flexible work environment and the chance to accelerate your career while influencing next-generation AI platforms.Microsoft's mission is to empower every person and every organization on the planet to achieve more. As employees we come together with a growth mindset, innovate to empower others, and collaborate to realize our shared goals. Each day we build on our values of respect, integrity, and accountability to create a culture of inclusion where everyone can thrive at work and beyond. #EiP
ResponsibilitiesDesign and implement algorithms that enable transparent, interactive AI agent workflows. Collaborate with engineers and researchers to integrate HITL capabilities, memory, knowledge, and evaluation frameworks into the platform. Analyze agent-human interactions to extract insights, optimize task prioritization, and create adaptive learning mechanisms. Drive experimentation, validate hypotheses, and contribute to scalable solutions that improve both internal and client-facing AI agents. | |