Mehdi Dastani

Mehdi Dastani is a computer scientist at Utrecht University who studies autonomous agents and multi-agent systems. After leaving Iran during its war with Iraq, he studied computer science and philosophy at the University of Amsterdam and earned his doctorate there in 1998. In Utrecht, he worked on the agent programming language 3APL with Frank Dignum and John-Jules Meyer, published its successor 2APL in 2008, and coauthored large-scale agent-based simulations of COVID-19 in Virginia from 2021 to 2023. His election as a EurAI Fellow came in 2024. Today he is professor and chair of the Intelligent Systems group, which he took over from Meyer in 2022, and program leader of Utrecht's master's programme in Artificial Intelligence.

Now at: Utrecht University
Born: 1964
Works: 578
Citations: 6,987
h-index: 38
Badges: Loyal

Years link to their sources. Publication numbers from OpenAlex, as of 22 October 2026.

Timeline

1985–1998

Computer science and philosophy in Amsterdam

As a high school student in the 1980s, he leaves Iran during its war with Iraq, and he reaches the Netherlands in 1985. He studies computer science and then philosophy at the University of Amsterdam, explores neural networks for his master's thesis in the early 1990s, and earns his doctorate there in 1998.
2001–2003

The 3APL agent language

He arrives in Utrecht as a postdoc from Amsterdam in 2001 and works on multi-agent systems in John-Jules Meyer's Intelligent Systems group. In April 2003, he, Frank Dignum and Meyer describe 3APL, a programming language for cognitive agents and the high-level control of cognitive robots, with an interpreter written in Java.
2008

2APL, a practical agent language

In 2008, he publishes "2APL: a practical agent programming language" in the International Journal of Autonomous Agents and Multi-Agent Systems. Developed at Utrecht University as the successor of 3APL, 2APL is a modular BDI-based language for building multi-agent systems.
November 2018

Professor of Intelligent Systems

In November 2018, Utrecht University appoints him Professor of Intelligent Systems in the Department of Information and Computing Sciences, where his research centers on autonomous systems and multi-agent systems. With the appointment, he becomes program leader of the master's programme in Artificial Intelligence and helps coordinate the bachelor's programme.
March 2019

A talk on machine emotions

In March 2019, he speaks at Utrecht's Academiegebouw on whether artificial intelligence can learn to be hopeful, joyful or in doubt, in the Studium Generale event "A different state of mind". His research covers what emotions are, how they can be modelled and what it means when AI systems show realistic behaviour towards people.
2021–2023

Simulating COVID-19 in Virginia

In 2021, he coauthors with Jan de Mooij and others a data-driven simulation of COVID-19 in the US state of Virginia, presented at the MABS 2021 workshop, in which agents reason about their beliefs, trust in government and the norms the government imposes. In 2023, the team publishes a framework in the journal SIMULATION that runs epidemic simulations with millions of deliberating agents.
2024

A EurAI Fellowship

He is elected a EurAI Fellow in 2024 for his contributions to formal and computational models for autonomous agents that interact with other agents.
As of October 2026

Chair of Intelligent Systems

He succeeds John-Jules Meyer as chair of the Intelligent Systems group in 2022. He is professor and chair of the group and program leader of Utrecht's master's programme in Artificial Intelligence, and he takes part in the Hybrid Intelligence Centre's project on monitoring and constraining adaptive systems. In May 2026, he coauthors an AAMAS paper on neuro-symbolic action masking for deep reinforcement learning.