Noryen - Monitor, compare and catch risky AI outputs

I was building another app in a sensitive niche (health) and at some point it hit me: my AI could randomly recommend medication, dosages or just say something completely unsafe… and I wouldn’t even know. This could lead to legal issues, broken user trust and unpredictable outputs. So I built Noryen in order to understand how my AI behaves in production Simple and built for those who don’t fully trust their AI.

Thanks everyone for checking out Noryen. 🙏  This came directly from a mistake I made.. While building Biometryx (health related app) I realized my AI could easily generate unsafe advice… and I had zero visibility unless I manually tested everything. So I built this tool. Nothing fancy, just what I needed as a developer. If anything feels off, confusing, or just not useful - tell me directly or ping me on X: https://x.com/StGrozdanov  I’m building this very actively and your feedback will shape what comes next. I Would love to hear: how you’re handling AI safety (if at all) or if you’re just trusting the model and hoping for the best 😄

Noryen - Monitor, compare and catch risky AI outputs

Replies