.png)
Course description
One major use-case for Large Language Models (LLMs) is assisting in, or taking over, writing computer code. To achieve this one can ask for computer code in a chatbot.
LLMs become much more powerful when used in combination with a so-called harness application. These provide a framework around the language model that governs how the model is used. Popular harnesses are Claude Code/Cowork, ChatGPT/Codex, GrokBot and Antigravity.
In this workshop, we will show how an open-source harness application called OpenCode (opencode.ai) can be used to enable the writing and running of code through the application assisted by an LLM. Use cases introduced in the workshop are data analysis, visualizations, managing files and folders, working with entire code bases, automation of tasks, and production of documents or webpages. We will use models available through GPT UiO (gpt.uio.no) through its API access.
While we teach using OpenCode, the material is equally applicable for commercial harnesses (Claude, ChatGPT, etc) and how to use these will be discussed.
The last few hours of the workshop participants will work on their own project or try out more of the things they have learned, with help from the instructors. We conclude with a show-and-tell of participant results.
Target audience
Researchers in any scientific field that work with data(files) and/or write computer programs (in any language).
Learning outcomes
- After this workshop, participants will be able to install OpenCode on a laptop or other computer
- Be able to configure the app to work with models made available through, for example, the GPT UiO API
- Be able to use OpenCode to chat with an LLM, set up and run simple data analysis workflows, create visualizations, managing files and folders production of documents or html-based pages
- Be able to use OpenCode for working with an entire code base and automation of tasks
- Know the benefits the iterative process to obtain a satisfying result
- Know how to be considerate of the AI regulations at UiO when planning LLM-assisted tasks
- Are aware of how other, similar tools, can be used with UiO provided LLMs
- Be able to discuss ethical and practical considerations when using LLM-based tools in scientific research
Prerequisites
- Entry level: beginner/intermediate.
- Experience with Python programming or a similar programming language is a benefit but not a requirement. The workshop will be programming language agnostic.
- Knowing the basics of Large Language Models (LLMs) and their usage is a benefit but not a requirement.
Required materials
- A laptop with either administrator access, or access to a remote compute instance such as EduCloud, or Sigma2 resources.
- Access to the GPT UiO API, we will request this for those that do not have it yet.
Lunch
- There will be a complementary buffet-lunch at 12:00-13:00 from the Sverdrups spiseri (Georg Sverdrups Hus, 2nd floor).
- After you get your lunch, join us at Store møterom where we have reserved seats for #DSDays2027 participants.
Number of participants
30
Contact
If you have any questions regarding this course, or anything else about the DSDays 2027 event, please contact us per email: digitalscholarship@ub.uio.no