API not giving the right outcomes

Where does the bug appear (feature/product)?

Cursor IDE

Describe the Bug

I am building my MVP called Skill Sight, which helps with parsing of marketing professional resumes and does a diagnosis of their AI skills gaps and recommends learning path with free youtube videos. I have added all the relevant API keys despite this the AI gaps and learning path videos being recommended are the same ones. I tried troubleshooting multiple ways with Cursor Ask and other LLM’s but I keep getting the same info. Need help

Steps to Reproduce

I am adding a LOOM video to show you the issue

Expected Behavior

The priority gaps and youtube videos are same for different users.

Screenshots / Screen Recordings

Operating System

MacOS

Current Cursor Version (Menu → About Cursor → Copy)

Version: 1.5.5 (Universal)
VSCode Version: 1.99.3
Commit: 823f58d4f60b795a6aefb9955933f3a2f0331d70
Date: 2025-08-25T17:40:25.290Z
Electron: 34.5.8
Chromium: 132.0.6834.210
Node.js: 20.19.1
V8: 13.2.152.41-electron.0
OS: Darwin arm64 24.6.0

For AI issues: which model did you use?

Model Name- Chat GPT 5

Does this stop you from using Cursor

No - Cursor works, but with this issue

Thanks for reporting this, @Ritesh_Bhatia! I want to make sure I understand the issue correctly so we can help you effectively.

From your description, it sounds like your Skill Sight application is producing the same AI skill gaps and YouTube video recommendations for different users, even though you’ve configured API keys.

To clarify:

  1. Is this an issue with your application’s code/logic (the app you’re building produces identical results for different resumes)?

  2. Or is this about Cursor not helping you debug the problem effectively?

If it’s #1 (application logic issue), here are some common debugging steps:

  • Check if there are hardcoded values in your code for the skill gaps or video recommendations

  • Verify your API calls are actually executing (add console logs before/after API calls)

  • Ensure resume data is being parsed correctly and passed to your recommendation logic

  • Check if there’s any caching or state management that might be reusing old results

  • Verify your YouTube API and other external API responses are actually different for different inputs

If it’s #2 (Cursor not helping), we can look into that separately, please share more details regarding that.

Super excited for you building your MVP with Cursor!

This topic was automatically closed 22 days after the last reply. New replies are no longer allowed.