Dock Vision AI
1st place IEEE Hackathon win, then delivered to a client.
- Client
- Confidential
- Year
- 2025
- Category
- Computer Vision
- Role
- Design + Build
- Timeline
- Hackathon build, then production hardening
The problem
Loading-dock turnaround is guesswork. Nobody knows exactly when a truck arrived, how long it sat, or where the bottleneck is, so the numbers that would fix scheduling simply do not exist.
What we built
A live operations dashboard driven by computer vision on the dock camera. It detects and tracks vehicles and forklifts through the bay, marks each state change (arrived, positioned, loading, then released) on a live turnaround timeline, and streams an event feed operations can watch in real time. Built for the hackathon, then delivered as a commercial product.
The result
Won first place at an IEEE Hackathon, then was delivered as a commercial product to an international client.
- vehicle + forklift detection on the bay camera
- real-timevehicle + forklift detection on the bay camera
- arrival → release timeline, logged automatically
- per truckarrival → release timeline, logged automatically
- IEEE Hackathon, then delivered to an international client
- 1stIEEE Hackathon, then delivered to an international client
Built with
- Python
- YOLO
- OpenCV
- Object tracking
- FastAPI
- React
We deliver what we commit.
Tell us what you're trying to build.
We'll come back within 24 hours with honest feedback on scope, timeline and cost, whether or not we turn out to be the right fit.