🤖 Agentic HR Assistant

Project Overview

This project demonstrates a functional Agentic AI designed to reduce HR administrative overhead. It utilizes a Retrieval-Augmented Generation (RAG) architecture to allow an LLM to "read" internal company policies and execute basic HR tasks autonomously.

System Architecture

The agent follows a Reasoning + Acting (ReAct) logic:

1. User Query: "How many vacation days do I have left?"
2. Reasoning: The LLM identifies it needs the check_vacation_balance tool.
3. Action: Python executes the script to query the database.
4. Observation: The script returns "12 days."
5. Response: The LLM generates a natural language answer for the employee.

---