Built this cuz no library gave LLM API calls in prod a consistent way to retry, timeout, circuit-break and cache across multiple providers.
Show HN: VernLLM – The AI resilience layer for TypeScript
vernllm.vercel.app