Per-refill
AI-verified meter readings
Automated
Logging replaces paper records
Flagged
Anomalies caught in real time
The challenge
Fleet fuel spend was recorded on paper slips and manual entries, making reconciliation slow and fuel theft or misreported readings almost impossible to detect after the fact.
Our approach
We built a React and shadcn/ui dashboard on a Python/Django backend, with an LLM layer that extracts meter and odometer readings from refill photos, cross-checks them against consumption history, and surfaces discrepancies to fleet managers as they happen.