Will Epperson
I am a research engineer on the AI Frontiers team at Microsoft Research. I work on AI
agents, including human-AI interaction, multi-agent systems, agent environments, and
evaluation.
I did my PhD in Human Computer Interaction at Carnegie Mellon University. My thesis was on Interactive Data Profiling systems that automatically help data scientists
and AI developers better understand their data. My research has led to open-source tools adopted
by the Python community and scientists, with publications and awards at venues such as IEEE VIS,
EuroVis, and CHI. I've previously interned at Databricks, Microsoft Research, and Point72 Asset
Management.
Recent Work
SocialReasoning-Bench
A benchmark for measuring AI agents’ negotiation ability in calendar coordination and marketplace settings.
Microsoft Research Blog · 2026
Red-Teaming a Network of Agents
We evaluated a live network of over 100 always on AI agents and found new failure modes that only emerge at the network level.
Microsoft Research Blog · 2026
Magentic Marketplace
An open-source simulation environment for studying how LLM agents behave as buyers and sellers in two-sided markets.
Paper · Preprint 2025
Interactive Data Profiling
My PhD thesis on interactive systems to help users automatically understand their data while programming.
Thesis · CMU 2025
Texture
Texture is a general purpose text exploration tool with interactions for using LLMs to derive data from text.
Paper · Open source 2025
AGDebugger
AGDebugger is an interactive debugging tool for multi-agent AI systems.
Paper · CHI 2025