Mehrnoosh Sameki

Mehrnoosh Sameki currently leads the Generative AI and Agents Evaluation Science, AI Governance, AI Redteaming, and Safety Evaluation efforts within Azure AI Foundry at Microsoft, shaping how customers assess and govern AI models, generative AI applications, and agentic systems. Her team focuses on building scalable, responsible evaluation and governance capabilities that are deeply integrated into the Azure AI ecosystem, with a strong emphasis on AI safety and security.

Mehrnoosh co-founded several open-source tools in the Trustworthy AI space—Fairlearn, Error Analysis, and the Responsible AI Toolbox—and contributed to InterpretML, advancing transparency and accountability in machine learning. She is honored to be recognized on the 2024 Brilliant Women in AI Ethics list.

Beyond her role at Microsoft, she serves as a curriculum developer and lecturer with Break Through Tech, helping drive underrepresented women toward meaningful careers in data science. Mehrnoosh earned her Ph.D. in Computer Science from Boston University and served as an Adjunct Assistant Professor from 2020 to 2024.