Fleet maintenance and service booking agent

The brief

Fleet managers juggle alerts, workshop slots and parts by phone. Build an agent that reads vehicle alerts or a driver's description of a problem, suggests the likely issue, checks workshop slots and parts through tools, and books the service only after the manager confirms. Good looks like: tools with permissions, a confirmation step, a test set of scripted cases, and success rate and cost per booking.

What you will build

1. Build a simulated fleet, workshop and parts system
2. Expose it to the agent as tools
3. Build the agent with LangGraph and a confirmation step
4. Test on 30 scripted alerts and descriptions
5. Measure correct bookings, wrong actions blocked and cost
6. Log every action

Free, open data to build it with

• AI4I 2020 predictive maintenance (alert signals) ↗ UCI Machine Learning Repository

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.