Apple rejected my app for UGC. So I built an API to fix it
Hey everyone,
A few months ago, I tried to launch an app with user-generated content (it was coaching AI btw). Apple rejected it. They cited Guideline 1.2, which requires a system to filter objectionable material.
I looked for an API to handle this. I found AWS Rekognition and Hive. They are built for huge companies. They have opaque pricing, complex SDKs, and take hours to set up. I just wanted a simple boolean response (safe: true) to pass my app review.
So I built Tabu. It is an image and video moderation API made for indie developers.
Here is how it works:
It uses TensorFlow.js and MobileNetV2.
It runs on a $6 DigitalOcean Droplet.
It processes images entirely in memory. It deletes them instantly. Nothing gets saved.
I added a React dashboard so you have a "human review" queue if an image gets flagged.
I am looking for some feedback from other solo founders or developers. If you are building an app with image uploads, I would love to know if the integration steps in my docs make sense.
The free tier includes 5,000 requests per month. You do not need a credit card.
The simple integration is the strongest part here. For indie developers, getting from “my app has UGC” to a usable moderation layer without a large SDK or complicated setup is a meaningful problem, and the human-review queue is a useful fallback when automation isn't enough.