Research and Publications 

MAGICS Research Areas

Design and Governance of Sociotechnical Systems

We study methods to integrate top-down intervention with bottom-up behavioral responses (human and AI agents) in sociotechnical systems. We also build frameworks that merge the design and goverannce of complex engineering systems

Multi-Agent Human-AI Systems

We study the dynamic of beheaivor in multi-agent AI and human systems in mixed cooperation and competition situations. We use AI as proxy for social behavior; also explore model the strategic behavior of LLM and RL agents.

Computational Methods for Policy Design

We build cuasal statistical and simulation model for scenario analysis in complex sociotechnical systems (healthcare, platform marketplaces, urban infrastructures) to quantify the effect of various inteventions and policies.

AI in Systems Engineering

We integrate AI in Systems Engineering in three ways: Studying the effect of AI integration in the system using causal inference; Use AI agents to create adaptive system governance methods; and use GenAI agents (LLMs) to create digital twins of complex sociotechnical systems.

Selected Recent Peer-Reviewed Publications