Talk summary / 7:25 min
AI as Tool User: Coding agents for more productive study and research
Download video · English captions
In this video
- 0:00 From ChatGPT to a desktop agent
- 0:40 What it did for him
- 1:33 How AI use developed
- 2:50 How an agent works
- 4:10 What an agent does that a person would
- 4:49 What it means for study and research
- 5:44 What keeps it working
- 6:32 How to check an agent’s work
In this lecture to MSc students, Dominik Lukeš describes his switch from ChatGPT to a desktop agent. On his computer he opens Codex or Claude Code, with the same models as before. He defines an agent as a model using tools in a loop to achieve complex tasks that require multiple steps, and shows what it did for him: a catalogue of 207 hours of radio recordings, a security audit of his machines and a completed expenses claim. The key to making it work over time is context, supplied through files, and his advice is to build in checking and double-checking, depending on how much the result matters.
You'll come away with
- How a chatbot differs from an agent, shown with a downloads-folder example in which the model asks for file sizes and the harness runs the command.
- What agents can do for study and research, from a Zotero collection built from an Elicit search to a quick experiment that gave evidence instead of a guess.
- How to give an agent context: an AGENTS.md file in the top folder and skills as folders with a SKILL.md file, and what data leaves your computer for each task.
Narration uses a synthetic voice. This video is an AI-assisted summary of the recorded lecture.