llmops-duke-aipi

Rubric for LLMOPs class

This document outlines the rubric used for this class.

Ten weekly demos

Look at the requirements for the final individual project and create a week by week schedule for the next 10 weeks with specific milestones.

For each week, create a weekly demo. You will create a screen recording of at least 30 seconds where you will talk about the progress made that week.

Watch the included lessons and make sure you cover the “Weekly demo delivery” question in addition to the progress on your project.

• Initial Post (50%)
  • Accurately describes the week’s progress on the project with a screen recording
• Followup Post (50%)
  • Responds to two other student posts and demonstrates judgement

[!IMPORTANT] Do not build the model yourself. Use the Llamafile project as a reference.

Final project delivery

You will deliver a project based on Large Language Models. For this class, you will use a Local Large Language Model application. These are the requirements for the project:

1. Has to be very well documented with a README.md that explains what the main purpose of the application is, how to set it up, and run it.
2. The documentation needs to have workable examples of the application.
3. It needs to have tests
4. The project should have automation with GitHub Actions that can produce a working container image. The resulting image should not include the LLM.

Grading breakdown

Assignment	Description	Points	Percentage
Weekly demo	Post video and comment on 2 other student posts	100	50
Final Project delivery	A fully working repository with documentation, tests, and an automated build	100	50

Grading Scale

Grade	Percentage
A	93-100
A-	90-92
B+	87-89
B	83-86
B-	80-82
C+	77-79
C	73-76
C-	70-72
F	0-69