About Us
Our Roots Are in Research. Our Future is AG2.
At AG2, we aren't just building software—we are advancing the frontier of agentic intelligence. Our work is grounded in the research that gave birth to AutoGen, the multi-agent architecture described in "AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation." Rather than treating agents as siloed molecules, AutoGen's core insight was that intelligence is emergent when agents coordinate, converse, and share context.
But the story didn't end there. As the agent paradigm matured, the original AutoGen architecture revealed key gaps: in observability, modularity, extensibility, and governance. That's where AG2 enters the scene. Led by the original AutoGen creators, AG2 is the next chapter: a community-first, research-minded framework that rethinks agent orchestration with rigor and responsibility.
We combine the best of experimentation and engineering. AG2 is not a product off the shelf; it's an evolving platform, built in the open, that enables unified workflows spanning LLMs, tools, APIs, and human interaction. We believe the future of AI isn't monolithic — it's composed of systems interacting in harmony. With deep roots in academic rigor and open-source practice, AG2 stands at the intersection of research and impact.
We're Hiring
Join us in shaping the future of agentic AI. We're looking for talented researchers, engineers, and builders who share our vision.
What We Stand For
Research-First
Every feature we build is grounded in rigorous research and academic principles, ensuring our platform advances the state of the art.
Community-Driven
AG2 is built in the open, with contributions from researchers and developers worldwide who share our vision for agentic AI.
Production-Ready
We bridge the gap between research and real-world deployment, providing the tools needed for enterprise-grade agent systems.