In this tutorial you will extract structured data, markdown chunks, and table and figure summaries from research papers, create embeddings and payloads for each markdown chunk, upsert them into a Qdrant collection, and then filter and query the resarch papers using a LangGraph agent.
Try out this example using this Colab Notebook

​
Create Complete and Accurate Qdrant Points with the Power of Tensorlake

Let’s set the context for this example, you will build a LangGraph agent for a real estate company to help track who has signed property documents, when they signed, and who still needs to sign. You’ll learn how to:
  • Use Tensorlake’s Python SDK
  • Extract structured data, markdown chunks, and figure and table summarization from research papers with unique reading order
  • Create robust points (embeddings and payloads) to upsert into a Qdrant Collection
  • Create a LangGraph agent that uses simple tools to answer questions like:
    • Does computer science education improve problem solving skills?
    • What are the key takeaways from papers by Bill Griswold?
    • Why is learning to code hard for kids?

​
Prerequisites

​
Build and test your LangGraph Agent

1

Setup

​
Next Steps: Improve your Qdrant Embeddings with Tensorlake

Start using Tensorlake today with 100 free credits upon sign up at cloud.tensorlake.ai. We’d love to see what you build with this, you can share with us or give us feedback in our Slack Community.