📄 Document Summarizer using Amazon Bedrock & Streamlit
(Built as part of the AI for Bharat – Powered by AWS Learning Program)
This project is a document summarization application originally developed during the AI for Bharat – Innovation Hack2Skill Program (Media Partner: YourStory). The code in this repository is based on the official workshop lab materials, and I have made additional improvements to:
UI layout
File upload handling
Folder structure
Documentation
Architecture diagrams
This repository is intended for learning, reference, and demonstration only.
During the AI for Bharat workshop, we used a temporary, free, pre-configured AWS environment where Amazon Bedrock usage was covered by AWS.
Outside the workshop environment (i.e., on your own AWS account), Amazon Bedrock API calls are NOT free. Running bedrock.converse() or generating summaries will incur charges.
📘 AWS Bedrock Pricing
🔗 https://aws.amazon.com/bedrock/pricing/
📘 Model Billing Details
🔗 https://docs.aws.amazon.com/bedrock/latest/userguide/pricing.html
To avoid unexpected charges:
Disable Bedrock model access
Set up AWS Budget Alerts
Or run the UI without calling the API
✨ Features
Upload PDF documents
Enter custom summarization instructions
AI-generated summaries from Amazon Bedrock
Clean, two-column Streamlit interface
Easy to run and extend
🧠 How It Works
-
Upload a PDF file
-
Streamlit reads the document and prompt
-
Backend sends bytes + instructions to Amazon Bedrock Converse API
-
Claude (Sonnet) analyzes content
-
Summary is generated and displayed
🧰 Tech Stack
Amazon Bedrock (Claude 3 Sonnet)
Converse API with Document-Chat
Python 3 & boto3
Streamlit
PDF byte processing
🏗 Architecture Diagram
(Add your architecture image here)
User → Streamlit App → summarization_lib.py → Amazon Bedrock → Summary Output
🖥 UI Preview
(Add your UI screenshot here) 
📁 Project Structure
document-summarizer/ │ ├── app/ │ ├── app.py # Streamlit UI │ └── summarization_lib.py # Bedrock logic │ ├── docs/ │ ├── Final UI.jpg │ └── Archietecture.png │ ├── requirements.txt ├── .gitignore └── README.md
⚠ Reminder: Running on your AWS account triggers real Bedrock charges.
pip install -r requirements.txt aws configure streamlit run app/app.py
🚀 Use Cases
Report summarization
Research paper insights
Meeting transcript summaries
Customer case analysis
Academic content simplification
📝 Attribution & Event Details
This project is derived from the Content Summarizer Lab taught during the:
AI for Bharat – Powered by AWS
Organized by: Innovation Hack2Skill Media Partner: YourStory
The workshop provided:
Guided labs
Preconfigured Bedrock access
Educational sample code
I used the workshop’s base implementation and added modifications to improve:
Folder structure
UI design
Documentation
Readability
Architecture diagrams
This repository is not an official AWS product—it is an educational reconstruction and enhancement based on the training materials.
⭐ Support
If this project helped you, consider giving it a star ⭐ on GitHub!