Tracking LLM Response Improvements: Tools and Solutions Explained

Hi - thanks. To answer your question, I'm interested in tracking improvement/regression in the LLM responses we receive as we optimize our prompt (and our approach to pulling relevant examples into the prompt) . I assume this would utilize your "evaluations" feature but it's unclear to me if that is something I should pursue within Phoenix or Arize Cloud (or some other product/solution you offer). thanks for your help! Also - I looked at the notebook you provided. It's helpful, but includes a broken link ("...For the full details on the OpenInferenceTracer and LangChainInstrumentor, see the integrations documentation."....that link results in a "page not found error."