Manufacturing quality-control teams, retail analytics companies, drone/satellite inspection startups, and medical imaging platforms frequently search to hire a freelance Computer Vision (YOLO, Segment Anything SAM 2, OpenCV, PyTorch, TensorRT) programmer. I am Aldian Fazrihady, a Senior Freelance AI Application Engineer and Full Stack Programmer.
Core Computer Vision & Visual AI Pipeline Capabilities
- Object Detection, Instance Segmentation & Pose Estimation: Fine-tuning Ultralytics YOLO, Meta SAM 2, Grounding DINO, and custom PyTorch vision models for defect detection, shelf inventory counting, license-plate/container ANPR, and PPE safety compliance.
- Real-Time RTSP Video Stream & Edge Deployment: Processing live RTSP/WebRTC security and industrial camera feeds on NVIDIA Jetson / TensorRT, DeepStream, or cloud GPU workers with low-latency React bounding-box dashboards.
- Hybrid Classical OpenCV + Vision LLM Verification: Combining sub-10ms OpenCV geometric/contour filters with Gemini 2.5 Pro / GPT-4o Vision reasoning for complex anomaly classification.
Frequently Asked Questions
How do we train an accurate custom object detector when we only have 200 labeled images?
I use zero-shot foundation models (Grounding DINO + SAM 2) to auto-label thousands of raw frames, curate edge cases in CVAT/Roboflow, and fine-tune compact YOLO models that run at 60+ FPS.
How do I hire you as a freelance Computer Vision programmer?
Hire me via Upwork.