Why I Still Build in a World Full of AI Tools. PromptCoder

1. Why did I do this
   I found this topic interesting.
   Initially, I was doing the project just for myself - as a tool to edit and develop code faster through prompts. But then an idea came to me: why not make it a public product? So it became a side project.
   Honestly, I didn't think I'd go this far.

2. What helped me get to this level
   My 6 years of experience working without ChatGPT and similar tools really helped the project.
   Why is this important? Because I really understand what the code does, how to analyze it, and I don't blindly rely on AI. This speeds up the process, helps me get into it faster, learn faster. It's important that the tool helps - but you yourself understand what's going on.

3. What's not perfect yet
   The project is not perfect. There are both minor and major flaws - those that require too many resources to fix for now.
   But I believe that I'll work everything out over time.
   The goals for the near future are:

• test the current version,
• collect feedback,
• fix bugs.

The UI is still basic - yes, the design can be improved, but now I don't want to clutter the product with unnecessary features. Focused on the essence.

4. I want to get feedback from you
   I know that there are problems - but they are mainly related to limited resources.
   Nevertheless, I think that everything is set up well now.
   The project is launched in testing mode - some features are cut for the sake of performance (temporarily).

I will be glad to hear:

• How do you like the idea?
• Is the interface convenient?
• What would you add?
• My fingers hurt, what should I do?

5. Competitors? Of course there are
   I understand that the project has competitors - they are better secured financially, their team is larger, and their resources are more extensive.
   But... why not try?

There were moments when I wanted to suspend work.
But I thought - I've already gone too far.
I decided to give myself one last push: April + May - that's +4000 lines of code, all manually, through real tasks.
Not as easy as it might seem. But my experience in real projects helped.

6. Immediate goals

1. Learning mode
   Will allow you to learn any language or profession. The structure is almost ready.
   Unfortunately, you can't just "feed everything to AI" here - there are too many nuances. ChatGPT can't take into account many human factors, you need to think, not just ask.
2. UI/UX improvements
   Even though this is not a priority now, a visual and UX step forward is definitely needed.
3. Complex browser-level tasks
   As soon as I close the basic features, I'll move on to really complex frontend/infrastructure things.

https://promptcoder.it.com

P.S.
I'm going to hold my fingers under cold water, feeling all the uselessness.
And my mother told me, "use ChatGPT more often."