I will build custom object detection and image recognition models with yolo

Need software that can see? I build computer vision systems that run on your data, not on a tidy demo dataset.

I have delivered production vision work commercially, including a card recognition system for a European client, and an automated animal behaviour tracking system that a research lab now runs daily.

What I build:

Object detection and counting: people, vehicles, products, defects, in images or video

Image classification into your own categories

OCR: text from documents, receipts, ID cards, meter readings

Segmentation: pixel-level masks for medical and industrial imaging

Video analytics: tracking, action recognition, anomaly detection

I pick the architecture to fit the problem rather than forcing one framework: YOLO and RT-DETR for real-time detection, Faster R-CNN when accuracy matters more than speed, Vision Transformers and ConvNeXt for classification, U-Net, Mask R-CNN and SAM for segmentation, all in PyTorch or TensorFlow.

You receive trained weights, documented Python source, a performance report with mAP, precision and recall, and inference results on held-out images so you can see how it performs before you deploy.