Hey Product Hunt! 👋 Haarvish here, co-founder of Ferronyx. The problem that drove us to build this: Every time a robot failed in production, our team would spend 3–4 hours SSHing into machines, grepping logs, checking which ROS2 topics stopped publishing, replaying bag files, and manually stitching together infrastructure metrics with ROS state — hoping to find the root cause. It was painful, repetitive, and completely avoidable. So we built Ferronyx — the intelligence layer for ROS/ROS2 fleets in production. Real-time monitoring of ROS2 nodes, topics & actions + infrastructure (CPU, GPU, memory, network), combined with AI-powered root cause analysis that automatically connects the dots for you. Our beta customers went from 3–4 hours of debugging to 12–15 minutes. We'd love to hear from anyone running robots in production — what's your biggest debugging headache? Drop a comment, we read every single one. 🤖"

Ferronyx

Intelligence layer for robots

Intelligence layer for robots