This proof-of-concept aims to develop a conversational agent to assist hospital staff. The selected individual will build a data pipeline in Azure Data Lake to organize relevant information from an existing PostgreSQL database into a machine learning format. A natural language model will then be trained on this dataset to power a chatbot with GPT-like abilities. The intelligence behind the agent will come from the constructed machine learning model. An API must also be developed connecting the trained model to the chatbot interface. This prototype seeks to demonstrate the value of a dialogue solution for clients by deriving insights from structured data and responding to queries similar to human conversation. Applicants should possess skills in data engineering, machine learning, software development and designing applied AI systems.
