Creating a RAG application with AWS CDK as IaC, Qdrant and LlamaIndex ☁️

This repository contains a full RAG application using AWS CDK as IaC, LlamaIndex framework, Qdrant as a vector database, and deployment on AWS using a FastAPI app and Dockerfile.

For detailed project descriptions, refer to this Medium article.

Main Steps

Data Ingestion: Load data to an s3 bucket
Indexing: Use SentenceSplitter for indexing in nodes
Embedding and Model: OpenAI
Vector Store: Use Qdrant for inserting metadata
FastAPI and AWS: Handle requests via the FastAPI app deployed on AWS
IaC: AWS CDK

Feel free to ⭐ and clone this repo 😉

Tech Stack

Project Structure

The project has been structured with the following files:

app: app logic with Dockerfile
tests: unittest
app.py: AWS CDK synthesizer
aws_cdk_rag_fargate: script to create the constructs and stack
scripts: scripts to puch the documents to s3 and create the Qdrant collection
.env: environmental variables
requirements.txt: project requirements

Project Set Up

The Python version used for this project is Python 3.11. You can follow along the medium article.

Create an empty respository locally. This is necessary to initialize CDK. Afterwards you can copy the files from this repository.
Initialize AWS CDK to create the project structure:
```
cdk init app --language python
```
Create the virtual environment named main-env using Conda with Python version 3.11:
```
conda create -n main-env python=3.11
conda activate main-env
```
Install the requirements.txt:
```
pip install -r requirements.txt
```

Make sure the .env file is complete:

 QDRANT_URL=
 QDRANT_API_KEY=
 DOCUMENT_BUCKET=
 OPENAI_API_KEY=

Synthesize the app:
```
cdk synth
```
Bootstrap the app to provision s3, ECR and IAM roles:
```
cdk bootstrap aws://{Account ID:}/{region}
```
Create Qdrant Collection by running the script under scripts:
```
python s3_uploader.py
python qdrant_setup.py
```
Deploy the app on AWS CDK
```
cdk deploy
```

Get the DNS Name

aws cloudformation describe-stacks --stack-name AwsCdkRagFargateStack --query "Stacks[0].Outputs[?        OutputKey=='LoadBalancerDNS'].OutputValue" --output text

# Output
AwsCdk-RagFa-a1TpSwcX5iaS-653394042.eu-central-1.elb.amazonaws.com

Send a POST request to test the app

curl -X POST -H "Content-Type: application/json" -d '{"question":"What is a transformer?"}' http://AwsCdk-RagFa-a1TpSwcX5iaS-653394042.eu-central-1.elb.amazonaws.com/query

Clean up
```
cdk destroy
```

benitomartin/aws-cdk-rag-fargate

Creating a RAG application with AWS CDK as IaC, Qdrant and LlamaIndex ☁️

Tech Stack

Project Structure

Project Set Up