I built an internal monitoring tool that connects to my AI app, and alerts me when I’m about to hit rate limits, so I can fallback as needed and prevent chat blackouts. Found it easier than tracking multiple usage pages across OpenAI, Anthropic, Google AI studio.
Wondering if this is an issue others have faced? Those of you running AI apps in production, how do you monitor rate limits across providers?