This hands-on workshop introduces researchers to using large language models (LLMs) programmatically to support qualitative research workflows. Using real New Zealand legislative data, you will practice prompting techniques to prepare, label, and analyse unstructured text and images — then learn how to critically evaluate and validate what the model produces. No prior coding or machine learning experience is required.
Click a notebook link below, sign in with your Google account, and run the cells from top to bottom.
| Notebook | Open in Colab |
|---|---|
| 01 — Environment setup |  |
| 02 — LLMs as research instruments | |
| 03 — Exploratory analysis | |
| 04 — Thematic analysis and validation | |
| 05 — Visual feature extraction | |
Developed by Dr Kyle Hemming, Centre for eResearch, University of Auckland