← Back to projects
Car Brand Detection API
This was an experiment to see if I could build and deploy a real-time AI service entirely on my own. The biggest challenge by far was training a YOLOv5 model on my personal laptop. With a limited dataset and hardware, the model's accuracy is not high enough for a production environment. However, i did engineering the complete pipeline: training the model, diagnosing its weaknesses, and deploying it as a stable REST API using FastAPI.
PythonFastAPIYOLOv5PyTorchComputer Vision
View source on GitHub →