> ## Documentation Index
> Fetch the complete documentation index at: https://docs.tensorlake.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Integrations

> Connect Tensorlake with your favorite AI frameworks, vector databases, and data platforms.

Tensorlake gives AI agents secure, stateful sandboxes and durable orchestration for long-running workflows. It plugs into the rest of your AI stack — agent frameworks, vector databases, and data platforms — so you can build end-to-end agentic workflows, including document understanding when a workflow needs it.

## Orchestration Frameworks

Run your agents in Tensorlake sandboxes and orchestration, and call Tensorlake tools — including document understanding — from your preferred framework.

<CardGroup cols={2}>
  <Card title="LangChain" icon="link" href="/integrations/langchain">
    Use Tensorlake loaders and tools directly within your LangChain chains and agents.
  </Card>

  <Card title="OpenAI Agents SDK" icon="openai" href="/examples/agentic-applications/deep-research">
    Power your OpenAI agents with structured data extraction and document processing tools.
  </Card>
</CardGroup>

## Vector Databases

Index your parsed documents efficiently for Retrieval Augmented Generation (RAG).

<CardGroup cols={2}>
  <Card title="Qdrant" icon="database" href="/integrations/qdrant">
    Upsert structured points and embeddings directly into Qdrant collections.
  </Card>

  <Card title="Chroma" icon="database" href="/integrations/chroma">
    Store and retrieve document chunks with rich metadata in ChromaDB.
  </Card>
</CardGroup>

## Data Platforms

Turn unstructured documents into analytical tables for SQL queries and business intelligence.

<CardGroup cols={2}>
  <Card title="Databricks" icon="chart-simple" href="/integrations/databricks">
    Load extracted data into Databricks Delta Tables for large-scale analytics.
  </Card>

  <Card title="MotherDuck" icon="duck" href="/integrations/motherduck">
    Query your parsed documents using SQL with DuckDB in the cloud.
  </Card>
</CardGroup>