Postdoctoral Researcher in Computational Social Science

Department of Ecology

The Swedish University of Agricultural Sciences (SLU) is one of Northern Europe largest academic hubs for ecological research and offers a dynamic and excellent research environment with modern infrastructure. The Department of Ecology has about 150 employees, of whom around 40 work at the Grimsö Research Station in Bergslagen. Together, the SLU Ecology Centre and Grimsö Research Station conduct research on sustainable agriculture and forestry, plant protection, nature conservation, and wildlife management, and provide scientific knowledge to inform Sweden and Europe's environmental policies.

About the position

Project

Can we foresee the outcomes of public policies, political choices and other decisions? We are finding out by developing serious games and running them using AI.
We are building large-scale simulations of societal processes (environmental negotiations, nature conservation policies, and hybrid-threat scenarios) in which every actor is an autonomous AI agent powered by Large Language Models (LLMs). These agents simulate real-world stakeholders, from government ministers to interest groups, and interact through natural language in complex strategic settings. We then run thousands of iterations to map the distribution of outcomes. This effort is part of a new research program Articulating Complexity ( https://www.slu.se/articulating-complexity/). We are looking for an ambitious post-doc to join this effort.

Tasks and duties

You are the person who makes the simulations meaningful. The technical infrastructure produces outputs in the form of thousands of LLM-generated texts from prompts. Your job is to ensure that these outputs are grounded in real human behavior, real political dynamics, and real institutional logic. Concretely this means:

• Scenario architecture. Each simulation is defined by a scenario with an initial world state, a cast of agents, communication channels, institutional constraints, document corpora. You will design scenarios focusing on negotiations of nature conservation policies and socio-ecological hybrid warfare.
• Agent persona construction. For each agent you will define the identity prompt, ideology vector, personality traits, decision mode weights, and the affect parameters that govern when agents escalates beyond their default moderation. This means extracting positions, rhetorical styles, strategic preferences, and psychological profiles from real-world sources.
• Analysis and validation of simulation outputs. After Monte Carlo runs, you will analyze the distribution of outcomes (e.g. which policies emerge) and will also design behavioural signature validation (e.g. verifying that simulated agents exhibit the expected action distributions when compared against empirical baselines).
• Publication and dissemination. You will co-author scientific publications from the project. The research will produce papers at the intersection of computational social science, ecology, security studies, and AI, a space with very few competitors and high visibility.

You will work closely with the PI (Guillaume Chapron https://www.slu.se/en/profilepages/c/guillaume-chapron/%20) and a research engineer in AI (recruited concurrently). The three of you will form the core team.

Your profile

Required

• A PhD in computational social science (or political science, psychology, behavioural economics, evolutionary anthropology or a related field) with strong quantitative skills. The PhD must be completed by the start date.
• Ability to reflect about strategic behaviour, political dynamics, or social processes in a structured, analytical way, whether through formal models, experiments or quantitative analysis.
• Willingness to work with AI and LLMs. You should be comfortable working critically with AI-generated outputs, interpreting how LLMs respond to instructions, and contributing to the design, testing, and refinement of prompts.
• Strong writing skills: clarity, precision, and the ability to communicate across disciplinary boundaries.

Desirable

• Experience with agent-based modelling, game theory, simulation, or experimental methods in the social sciences.
• Experience with LLMs and their capabilities for simulating human behaviour such as the emerging literature on "silicon samples" and next-generation social simulations.
• Understanding of institutional design, legislative procedures, or multi-level governance. The simulation enforces procedural constraints, and designing these correctly requires knowing how real institutions actually work.
• Reading proficiency in Swedish, not required but an advantage.

Assessment criteria

• Intellectual depth and originality. We are looking for someone who has demonstrated the ability to produce genuinely nove