I’m a 3rd year CS student and have been building internet apps from home since I was 15.

I’ve recently been hired as an engineer at a startup and had this recurring problem of not being able to effectively test the use of my system prompts on different language models, and see the difference in output and pricing.

I needed it for work, to be able to show evidence of an LM provider choice being the correct one.

As an engineer, I want to know I’m choosing the right LM for each task I deploy on an application.

So GlidePrompt takes an arbitrary task -> creates system prompt -> lets you message the popular open and closed source LMs.

Let me know what you think so far and what direction you want to see this going in. Thanks for the feedback, positive or negative.