I built PromptProbe after getting burned by inconsistent AI outputs
I kept running into the same problem while building AI workflows: a prompt would work once, then behave differently on the next run.
Existing evals helped, but I wanted a faster way to spot output drift before shipping changes to production.
So I built PromptProbe. It runs the same prompt multiple times, compares outputs, and highlights where responses diverge.
I'm still figuring out where it fits in the AI testing stack, but the conversations so far have been surprisingly insightful.
This is clever. The 'prompt comparison' feature is exactly what people don't realize they need until they see two outputs side by side.
One thing though — tools like this spread fastest inside tech communities where people actively share AI tips. I've seen AI tools blow up just from being posted in the right WhatsApp and Telegram groups.
I run a launch distribution service for founders. I post your product into 20-50 active tech/AI communities on WhatsApp and Telegram — places where early adopters actually try new tools and give feedback. You get screenshot proof of every post, and you pay only after you see the proof.
$30 for 20 communities, $60 for 50. I can do 5 communities free first if you want to test the water.
Either way, good luck with this. The landing page is clean."