This chatbot will be trained on a diverse dataset that Amazon has accumulated over time. It will process user inputs, including text, images, and voice, to provide personalized product recommendations. The chatbot will have a user-friendly interface for seamless interaction. We leverage the EmbedChain framework, a Python-based tool for creating powerful bots powered by large language models. AWS services will be integrated for robust backend support, ensuring scalability and efficiency.

Backend service for LLM AMY

POST /query

{
  "query": "mero ko ek 5g phone chaiye jisme AMOLED display ho",
  "device": "Android/iOS",
  "language": "English"
}

{
  "status": true,
  "response": "PRODUCT_IDs: 1289"
}

1. Build the docker image with the following command

   sudo docker build -t amazon_hackon_amy_backend .

2. Run the docker image

   sudo docker run -dit --rm -p 4321:4321 amazon_hackon_amy_backend

3. The service will be live on localhost:4321

1. Setup the virtual environment

   python -m venv .venv

2. Activate the virtual environment

   source /.venv/bin/activate

3. Install all the dependencies

   pip install -r requirements.txt

4. Start the application

   uvicorn main:app --host 0.0.0.0 --port 4321

5. The application will be live on localhost:4321
6. If you encounter any error, try using Python 3.10.0

1. The response of the model is strictly based on the training dataset (only 2000 products)
2. The LLM model can cause hallucinations sometimes