Intelligent Document Processing with Bedrock Data Automation and AgentCore

This notebook demonstrates how to create an intelligent document processing (IDP) solution using Amazon Bedrock Data Automation (BDA) with Amazon Bedrock AgentCore Runtime. Instead of traditional Bedrock Agents, we'll deploy a Strands Agent through AgentCore, providing enterprise-grade capabilities with framework flexibility.

Note: This is sample code and not intended for production use

The workflow guides you through:

• Creating a Knowledge Base with BDA for multi-modal document processing
• Building a Strands Agent that integrates with the Knowledge Base
• Deploying the agent to AgentCore Runtime for production use
• Testing the deployed agent with complex academic report analysis

Agent Architecture

In this example we will create an academic performance assistant agent that connects with a Knowledge Base for Amazon Bedrock containing the academic reports.

Running the Notebook

1. Create and activate a virtual environment

python -m venv .venv
source .venv/bin/activate

2. Install dependencies

pip install -r requirements.txt

3. Register your virtual environment as a kernel for Jupyter

python -m ipykernel install --user --name=notebook-venv --display-name="Python (notebook-venv)"

4. You can list your kernels using:

jupyter kernelspec list

5. Launch the notebook

jupyter lab bedrock-data-automation-with-agents.ipynb

Important: Select the correct kernel by going to Kernel → Change kernel → "Python (notebook-venv)" to ensure your virtual environment packages are available.

Prerequisites

• An AWS account with credentials configured (aws configure)
• Python 3.10 or later
• Access to Amazon Bedrock models (Nova Pro, Claude 3.5 Sonnet)
• JupyterLab or Jupyter Notebook installed

Security

See CONTRIBUTING for more information.

License

This library is licensed under the MIT-0 License. See the LICENSE file.