Show HN: AI Assistant to Help You Learn Programming
profound.academy
Show HN: AI Assistant to Help You Learn Programming
1–6 of 6 posts
Re: Show HN: AI Assistant to Help You Learn Programming
#2Learning with a personal assistant is the future of computer science education! Imagine having a knowledgeable guide to help you every step of the learning journey.
As you work through our courses, the AI assistant will be there to analyze your code, catch mistakes, and provide actionable feedback and suggestions. With real-time guidance, you can streamline your learning process, save time, and gain a deeper understanding of the subject matter.
Check out the demo (in the link) and see for yourself the power and potential of learning with a personal AI assistant.
Re: Show HN: AI Assistant to Help You Learn Programming
#3I love how ChatGPT can be used to make useful features like this, great work!
Re: Show HN: AI Assistant to Help You Learn Programming
#4I love how ChatGPT can be used to make useful features like this, great work!
Thanks! We're currently using the GPT 3.5 version with some hacks to make it produce good outputs.
It would be great to try this with the ChatGPT as soon as they release a public API
Re: Show HN: AI Assistant to Help You Learn Programming
#5I wish I'd had this tech, when I started long ago. I'm curious: is your actionable feedback always going to give an answer / correct the code?
E.g. What's the line between prompting GPT3 (assumption) to correct the code vs. explain the error vs. explain how to think about the problem.
Re: Show HN: AI Assistant to Help You Learn Programming
#6I wish I'd had this tech, when I started long ago. I'm curious: is your actionable feedback always going to give an answer / correct the code? E.g. What's the line between prompting GPT3 (assumption) to correct the code vs. explain the error vs. explain how to think about the problem.
We did a couple of tricks to get good suggestions from the model (although some of them were changed later):
1. We tried to generate several explanation + code samples with the model. After which the code was parsed and evaluated with the judging system. Finally, only the code that would yield a correct result after the evaluation would be shown to the user. This was actually modified in the latest release we made.
2. We prompt different types of responses - suggestions on how to modify the code, suggestions of completely new ideas, and suggestions on which parts of the code to fix exactly.