TerpWhisper

TerpWhisper

Tired of reading individual reviews for professors on PlanetTerp? TerpWhisper summarizes professor reviews and gives you the info you need to succeed.

Like

Comment

Main UI
UI of Search

Inspiration

We were frustrated at how long it takes for students to navigate and understand information from PlanetTerp

How we built it

Platform: Chrome Extension Manifest V3
Languages: JavaScript (ES6+), HTML5, CSS3 (Glassmorphism & Radial Gradients)
AI Model: gemini-3.8-flash via Google Generative Language API
Data Source: PlanetTerp API

Challenges we ran into

Gemini API rate limits
Parsing data we receive from the PlanetTerp API and Gemini
File redirection - how to navigate through the user input and Gemini output

What we learned

HTML and JS functions
How to use Gemini API in our project

What's next for TerpWhisper

Compare with ratemyprofessor.com reviews
Accessibility features and database management (e.g. ElevenLabs for audio)
Implement other tools provided by PlanetTerp

Built With

Submitted to

Technica 2026

Created by

I worked on connecting the input and output from the extension and Gemini API, and set up the GitHub workflows.

Abigail Chen
I take the raw PlanetTerp download and turn it into the file the rest of the project uses. When someone looks up a professor, PlanetTerp sends back every review in one big dump. I organize that into check.json, which has the professor’s name, overall rating, review count, an average for each course, and the newest reviews with the date, course, stars, and full text. This file is what goes to Gemini.

Martin Shijo
ariasongdc-cpu

Updates

Abigail Chen started this project — Oct 18, 2026 11:56 AM EDT

Leave feedback in the comments!

Log in or sign up for Devpost to join the conversation.