AI · Project
Lucid Motors AI service scheduling
I led product on an AI service-scheduling system at Lucid — it takes a customer's plain-English description of what is wrong with the car, predicts how long the repair will take, and books the appointment around the technicians actually available.
We had no UX designer at the time, so I stepped in and modelled out the metrics myself, prototyping this dashboard in Claude Code. The engineering team took the frontend more or less as-is into the product. It is not proprietary — it is simply how the team shares service metrics — so I am able to show the prototype here.
Note: this is my original prototype. The shipped product has evolved since; numbers shown are illustrative sample data.
Service Scheduling Engine Dashboard
Technician Utilization
83%
↑ 12% from last month
Next Appointment
2.4 days
↓ 1.6 days from baseline
FRT Prediction Accuracy
87%
↑ 5% from last month
Vehicles Aging >10 Days
10%
↓ 18% from baseline
Weekly Capacity Planning
Technician Utilization Trend
FRT Prediction vs. Actual
Work In Progress (WIP) Distribution
Technician Resources
| Technician | Skill Level | Mobile Tech | Available | Booked | Utilization |
|---|---|---|---|---|---|
| John Doe | 3 | Yes | 32 hrs | 28 hrs | 88% |
| Jane Smith | 2 | No | 40 hrs | 36 hrs | 90% |
| Mark Johnson | 3 | Yes | 40 hrs | 34 hrs | 85% |
| Lisa Brown | 1 | No | 32 hrs | 30 hrs | 94% |
| Robert Wilson | 2 | Yes | 40 hrs | 38 hrs | 95% |
Aging Vehicles Alert
Andrew Chen
Waiting for parts - Control module
Maria Rodriguez
Service approval pending
James Wilson
Complex diagnosis ongoing