Qingyun Wu

Founder & CEO, AG2ai, Inc. · Assistant Professor, Penn State University

I am the Founder and CEO of AG2ai, Inc. and co-creator of AutoGen (AG2), an open-source framework for building multi-agent AI applications. My work focuses on building practical systems that enable LLMs to collaborate, reason, and solve complex tasks through multi-agent conversation.

I am also an Assistant Professor at Penn State IST, where I lead research on AI agents, multi-agent systems, and automated machine learning. Before joining Penn State, I was a postdoctoral researcher at Microsoft Research. I received my Ph.D. in Computer Science from the University of Virginia in 2020.

Google Scholar GitHub

Research Interests

Multi-Agent Systems

Designing frameworks for LLM-based agents to collaborate through structured conversations and workflows.

AI Agents

Building autonomous and semi-autonomous AI agents that can reason, plan, and use tools to solve complex tasks.

Automated Machine Learning

Developing efficient and accessible AutoML systems for hyperparameter tuning and model selection.

Interactive Online Learning

Reinforcement learning, bandit algorithms, and sequential decision-making for real-world applications.

Featured Projects

AutoGen / AG2

The most widely adopted multi-agent framework — 50k+ stars on GitHub, 10M+ downloads, and a 21k-member Discord community.

View project →

FLAML

Fast, lightweight AutoML that powers Microsoft Fabric. 15M+ downloads and counting.