An automated triage system for Customer Support teams. This application uses a Hybrid AI Architecture (Machine Learning + Rule-Based Logic) to classify tickets, detect urgency, and predict customer churn risk in real-time.

Automated Classification: Uses ML to categorize tickets (e.g., Technical Support, Billing, Hardware Issues).

Smart Priority Detection: Combines Machine Learning with a Rule-Based Engine to catch critical issues (e.g., "System Crash", "Hacked") instantly, ensuring 100% compliance on safety-critical tickets.

Churn Risk Assessment: Analyzes customer sentiment (VADER) and ticket metadata to predict the likelihood of the customer leaving.

Guardrail System: Intelligent logic to detect "Positive Feedback" or "Routine Requests" (like "Send Invoice") and bypass unnecessary alarms.

Clean Architecture: Built using a Pipeline Pattern (Facade Design Pattern) to completely separate Business Logic from the UI.

Frontend: Streamlit

Backend: Python 3.8+

Machine Learning: Scikit-Learn, Pandas, Numpy

NLP: NLTK (VADER Sentiment Analysis)

Logging: Custom Logger & Exception Handling