MAS-Lab Open-Source Release and arXiv Paper
2026-06-30
We have released MAS-Lab, a specification-driven validation framework for reliable multi-agent systems, as open source on GitHub, together with an arXiv preprint.
Working at the intersection of systems research and engineering — from networking and security to agentic AI. This site gathers my research work and a personal blog where I document whatever I'm building or tinkering with at the time.
Current research
Design and evaluation of robust agentic AI systems
My current work focuses on the design and evaluation of robust agentic AI systems, with a focus on observability and interpretability. As part of this effort, I am chairing the Accuracy and Reliability working group of the AAIF.