Support ticket router

The brief

A software company's support inbox is triaged by hand every morning. Build a model that reads each ticket, assigns it to the right queue, sets its priority and extracts the product named, so tickets reach the right engineer first time. Good looks like: accuracy per queue and priority, a comparison of a classic model with a transformer, and a routing demo.

What you will build

1. Explore the tickets, queues and priorities
2. Build a TF-IDF baseline classifier
3. Fine-tune a small transformer and compare
4. Extract product names from the text
5. Show the routing in a small demo with confidence scores

Free, open data to build it with

• Customer support tickets ↗ Hugging Face

Check each dataset's licence before you publish your work. Your mentor gives your team its own version of the brief (data, scope or region), so every team's project is its own.