Aritificial Intelligence-based tools are increasingly popular. You probably used tools such as OpenAI ChatGPT and Google Bard for various purposes, but can you use them successfully in the context of High-Performance Computing (HPC)?

Learning outcomes

When you complete this training you will

• understand how Large Language Models (LLMs) work and understand their strong and weak points;
• understand that such tools are only useful if you already have a reasonable level of domain knowledge;
• understand various ways in which these tools can be used successfully;
• know what to expect from shell-gpt;
• know how to use ChatGPT and GitHub Copilot to generate tests for your code;
• know how to use ChatGPT and GitHub Copilot to generate API documentation for your code;
• use GitHub Copilot to generate boring, boilerplate code;
• use ChatGPT and GitHub Copilot for debugging;
• use ChatGPT Advanced Data Analysis for data exploration and visualization;
• use ChatGPT from the command line;
• use Ansible LightSpeed to generate Infrastructure-as-Code.

Schedule

Total duration: 2.5 hours.

Subject	Duration
introduction and motivation	5 min.
understanding LLMs	15 min.
code generation	15 min.
test generation	15 min.
documentation generation	15 min.
debugging and refactoring	15 min.
shell-GPT	15 min.
Ansible LightSpeed	30 min.
There be dragons!	30 min.
wrap up	5 min.

Training materials

Slides are available in the GitHub repository, as well as example code and hands-on material.

Target audience

This training is for you if you want to save time and use HPC systems more efficiently.

Prerequisites

You will need experience programming in some programming language for the code generation parts. You are also expected to be familiar with Linux and an HPC environment.

Trainer(s)

• Frederik De Ceuster (KU Leuven)
• Frédéric Wautelet (University of Namur)
• Geert Jan Bex (Hasselt University)