MongoDB Atlas Integration for Retrieval-Augmented Generation (RAG)

Guide to using MongoDB Atlas Vector Search with LangGraph agents logging to Galileo

MongoDB + RAG + Galileo overview

MongoDB Atlas Vector Search lets you keep your knowledge base inside MongoDB while enjoying hybrid search and AI-ready workloads. In this guide you’ll:

1. Embed and store vectors in Atlas
2. Setup Galileo
3. Stream LangGraph traces to Galileo for end-to-end observability

Set up MongoDB

Sign up to MongoDB Atlas and set up your credentials. You can setup the vector store as follows:

from langchain_mongodb import MongoDBAtlasVectorSearch
from langchain_openai import OpenAIEmbeddings
from pymongo import MongoClient

client          = MongoClient(MONGODB_URI)
DB_NAME         = "rag_demo"
COLLECTION_NAME = "blog_vectors"

collection = client[DB_NAME][COLLECTION_NAME]
embeddings = OpenAIEmbeddings()

vector_store = MongoDBAtlasVectorSearch(
    collection    = collection,
    embedding     = embeddings,
    index_name    = "rag-index",
    relevance_score_fn = "cosine"
)

Use Galileo for logging

import os
from galileo.handlers.langchain import GalileoCallback

# --- Galileo ---
os.environ["GALILEO_API_KEY"] = "<galileo-key>"
os.environ["GALILEO_PROJECT"] = "<galileo-project>"
os.environ["GALILEO_LOG_STREAM"] = "mongo_rag_demo"

# You can leave this commented out if you are using app.galileo.ai
# os.environ["GALILEO_CONSOLE_URL"] = "<galileo-console-url>"

galileo_handler = GalileoCallback()

Once the callback is created we can add it to the agent.

query = "What types of agent memory does Lilian Weng describe?"
inputs = {"messages": [HumanMessage(content=query)]}

result = graph.invoke(
    inputs,
    config={"recursion_limit": 5, "callbacks": [galileo_handler]}
)

Find the full notebook in Google Colab.