AI Agents in Economics and Business Research

Dates: May 13-14, 2027

Faculty 

Finn-Ole Höner is a PhD candidate for AI in Marketing at the Erasmus School of Economics. His work focuses on the conceptualization of how AI agents can be used to foster innovation in retail settings, the investigation of LLMs as tools for market research, and the development of new AI models that advance marketing science. 

With a background spanning computer science, data science, and econometrics, he bridges technical and marketing disciplines in both his teaching and research. He has taught in courses on text analytics, natural language processing, and machine learning, combining hands-on programming expertise with a deep understanding of data-driven decision-making and AI.

Course

Within this two day workshop, participants learn how to use current AI tools throughout the research workflow: generating and refining research ideas, reviewing and synthesising literature, analysing data and code, writing and revising, and communicating and disseminating research after publication. Day 1 focuses on practical use of off-the-shelf systems, including ChatGPT and standard agent harnesses such as Claude, Codex, or GitHub Copilot. Day 2 develops a deeper technical understanding of how AI agents work and how researchers can build and customise them using tools such as pi-agent and LangChain. The workshop combines demonstrations, guided exercises, and discussion of robust and responsible research practice.

Learning Objectives

Participants will learn to

• use LLM–based AI agents to tackle complex research and managerial problems in areas such as market research, ideation, and data analysis;
• apply core techniques in context engineering and multi-agent design, while paying attention to responsible use, and sustainability considerations;
• understand the basics of inner workings of language, image, video, and world-model AI systems;
• build hands-on workflows that integrate LLMs into everyday analytical and programming tasks, with attention to reliability, bias, environmental impact, and practical debugging.

Topics

• AI across the research workflow: ideation, literature work, analysis, writing, and dissemination
• Using off-the-shelf AI systems: ChatGPT, Claude, and Codex
• AI agents: concepts, architectures, workflows, and context engineering
• Building and customising research agents with pi and LangChain, including evaluation and responsible use

Level

The workshop welcomes PhD students, post-docs, TI & BDS Fellows, as well as practitioners who are interested in AI models.