Carousel Mini-Module: Drafting Letters of Recommendation
In this mini-module, participants will learn how to draft letters of recommendation using AI. This choose-your-own-adventure module allows participants to either use Google Notebook LM to generate personalized letters based on deidentified previous letters or use any text-generating AI to create a generic letter that can be customized. This module is particularly useful for those looking to save time while ensuring the letters remain authentic and personalized.
Bonus Introduction:
If you are not interested in the podcast, go to the Vanilla Intro below.
A Google NotebookLM podcast generated from the Vanilla Intro.
Vanilla Introduction:
Below you'll find a mini-module about drafing letters of recommendation using AI. This mini-module is a choose your own adventure module with three options! You can use Google Notebook LM and upload a bank of your previous letters of recommendation that have been deidentified (student last names and contact information removed), so the AI generates a letter that sounds like you. OR you can use any text generating AI and have it generate a generic letter of recommendation that you may need to spend more time making sound like you. OR you can use AI to generate a letter of recommendation template for students performing at different levels in your core courses that you can update with specific student information. I'll be honest. I prefer the output from Google NotebookLM. I didn't need to make as many revisions to make it sound like me. Corey's template is a great strategy if you have a lot of students from similar courses needing letters. Watch the videos with all three methods to help you decide which to try!
1. Jen's Video Using NotebookLM:
2. Jen's Video Using GenAI:
3. Corey's Video Creating AI to a Reusable Template:
4. Materials and Output:
Notebook LM
Chat GPT
Template Letter
- See the first prompt in Faculty Prompts Administrative [PDF]
5. Tool/Platform:
Bonus Step:
Try it at least two of the ways to compare and contrast the results. Which do you prefer? Why?