Workshop on Local Open AI

Do you have research or teaching tasks that are too large or repetitive to do by hand? Imagine analyzing a thousand articles for a literature review, classifying thousands of open-ended survey responses in a spreadsheet, identifying key themes across hundreds of student discussion posts, or generating custom study materials for your class, all in a matter of hours. This workshop introduces you to the power of local generative AI, a new library service that allows you to run powerful models securely on your own computer, keeping your research and student data completely private and cost-free.

New Library Service

This new service is called Academic Support Services for Local AI.

Part 1

In the first part of this hands-on session, we will guide you through the fundamentals. You will learn how to:

• Choose the right AI model for your specific task, considering your computer's hardware.
• Write effective prompts to get the exact output you need, including structured data like tables or JSON.
• Fine-tune model settings like temperature and top_p to control for creativity or consistency.
• Customize pre-made scripts to automate your unique research and teaching workflows.

Part 2

In the second part, we will demonstrate how to scale up these tasks with larger datasets using the university's High Performance and Research Computing Services (HPRC).

Open to All

This workshop is designed for all faculty, students, and staff. No prior experience in coding or AI is required. Join us to discover how you can integrate these powerful tools into your work.