This software contains an agent based on LangGraph & LangChain for solving general client queries and it could be implemented in whatever channel of this medical clinic (Whatsapp, Telegram, Instagram, etc).
It is highly autonomous and it is able to handle different types of secretary tasks like: give general information about the clinic, cancel, reschedule and set appointments, check doctor availability, review if your results are ready, which services the dental clinic offers, etc.
This is a casual chat where the agent books and reschedule books easily:
chat.mp4
Here I provided a wrong ID number and we can see how it handle errors quite good. Also, I provided a demo of how it handle general questions with a RAG approach.
error_handling_and_rag.mp4
- Set the following ENV variables
WORKDIR (The root path to the repository) XXX_API_KEY (LLM provider you want to use) PINECONE_API_KEY
-
If it is your first time, execute the main.py file inside the vector_database directory. This will create the Vector Database.
-
Execute the get_availability.py file inside the syntetic_data directory in order to have data up to date
-
Run the app executing agent.py / Run the app using LangGraph Studio