Building Odd Eye.
For vibe coders whose apps fall apart every time they add a feature.
Map it before the AI touches code. Solo founder, building in public.
Odd Eye probably isn’t for you if:
1. You want to skip the thinking and get something on screen as fast as possible.
2. You want AI to decide how your app should work.
3. You don’t want to review what the AI understood.
It may be for you if:
1. You’ve built a real app with Cursor, Lovable, or Bolt and learned how painful changes and maintenance can get.
2. You’re starting a new app or moving on to your next project.
3. You may not write code, but you still want to decide what gets built.
Odd Eye is built to put control of your app back in your hands.
AI organizes the messy details.
You review them and make the decisions.
Those decisions don’t disappear into a chat log.
They become a reference your coding AI can use during the first build and later changes.
That means less re-explaining, less rework caused by misunderstandings, and less chance of one change quietly affecting something else.
Describe your app in plain language.
Odd Eye turns it into something you can review, then creates a reference file for Cursor, Lovable, or Bolt.
AI organizes. You decide.
https://t.co/IlSEsp8fSp
#IndieHacker #BuildInPublic #SoloDev #Cursor #AI #DeveloperTools #SaaS
I’ve decided to discontinue the service I’m currently building.
Hi everyone, I’ve decided to shut down the "oddeye" service I’ve been working on.
There were two main reasons for this:
Uncertainty regarding exactly whose problem this feature solves, and
The fact that the costs far outweigh the benefits.
My goal was to enable non-developers—people who don't understand code—to comprehend and modify backend and frontend logic.
But... after using it myself, I realized that while the logic itself is understandable, a lot of the backend details are hidden.
Plus, the modification agent I built to allow for logic changes turned out to be quite expensive to run (costing around $0.90 per request).
For that reason, I’m going to pause the service itself, but I plan to keep working on the development little by little (I need to secure the funding to build it).
I’ll also drop by occasionally to share updates on the development process!
A huge thank you to everyone who followed me.
I still believe my approach represents the future of "vibe coding," and I’m going to come back and prove it.
Odd Eye Day 4
I built Odd Eye myself, but I still didn’t fully understand how it worked.
So I’m building a code translator that turns each piece of code into a plain-language sentence, so I can analyze Odd Eye properly.
It’s not finished yet, but I think I’ll have the first result tomorrow. Right now, I’m working on splitting the code into sentence-sized units.
I’ll share the result when it’s ready.
#buildinpublic #vibecoding #indiedev
@TechPriest_001 That’s right! As long as you set the right course and keep moving forward, you’ll get there eventually.
Even if the path is uncharted, the destination remains the same.
@luis_enriquez98 Well, I don't really pay much attention to the project name at first.
Since the app's identity emerges as I build it, the name inevitably ends up changing anyway.