RAG & Knowledge Basesfree

Evaluation metric example: RAG document relevance

by David RobertsUpdated Aug 2026

Requires

Google SheetsOpenAI

1/5

FLOWS

STEPS · 11

Starts on a Evaluation event

On an evaluation trigger, runs RAG queries over an in-memory vector store, scores document relevance with OpenAI, and logs metrics to Google Sheets.

Drive Knowledge Base Chat Agent

When files are added to Google Drive, indexes them into Milvus with Cohere embeddings so an OpenAI chat agent answers questions.

On a chat message, a Gemini agent answers using MongoDB Atlas vector search, OpenAI embeddings, and MongoDB chat memory.

On a Telegram message, runs retrieval Q&A over a Pinecone store using OpenAI embeddings and Groq, then replies in Telegram.