Workshop "AI-Assisted Empirical Research with Claude Code"
- https://www.uni-giessen.de/de/fbz/zentren/ggs/veranstaltungen/index_html/wise2026/claude-code
- Workshop "AI-Assisted Empirical Research with Claude Code"
- 2026-10-15T09:00:00+02:00
- 2026-10-23T17:00:00+02:00
15.10.2026 09:00 bis 23.10.2026 17:00 (Europe/Berlin / UTC200)
Licher Straße 68, Seminarraum 050 (vorläufig)
| Instructor: | Dr Florian Gärtner | |
| Dates: | October 15, 16, 22, and 23, 2026, 9.00 am - 5.00 pm | |
| Max. participants: | 15 | |
| Course language: | English | |
| Registration Deadline: | October 5, 2026 | |
| ECTS: | 3 ECTS |
Objectives
AI coding assistants such as Claude Code can now help with the whole empirical research pipeline, from cleaning data and writing analysis code to producing tables, figures, and prose. This workshop is a practical, hands-on introduction for researchers who want to use such tools seriously and keep control of their work. Rather than a tour of features, it builds the understanding needed to use an AI assistant deliberately, to decide what to delegate and what to keep, and to adapt as the tools change.
The workshop treats an AI assistant as a research collaborator to think alongside, not a machine to hand tasks to and trust without checking. Participants learn to brief and review it as they would a junior co-author, to verify its output, and to use it responsibly with respect to data protection and good scientific practice.
After the workshop, participants will be able to:
- explain how a large language model works, well enough to predict and diagnose its behaviour (why it “forgets”, why a long, cluttered context hurts performance, why it fabricates);
- set up and structure a research project so an AI assistant can navigate and work in it effectively (project conventions, version control);
- run an AI-assisted analysis pipeline end to end, from raw data to publication-ready tables and figures in LaTeX/Overleaf;
- configure Claude's behaviour to their own workflow (rules, permissions, hooks, custom commands, skills, agents);
- critically verify AI-generated results and recognise characteristic failure modes before they reach a paper;
- work with Claude as a research collaborator, including coordinating several parallel sessions safely (branches, worktrees);
- apply AI to literature work (including retrieval over their own library) and design their own multi-step research pipelines;
- integrate all of this responsibly, with respect to data protection and good scientific practice.
...
You can find more details in the syllabus "AI-Assisted Empirical Research with Claude Code".