Computer Vision Engineer
TeamPlus
3-7 Years
| • Experience with OCR engines and document understanding pipelines (Tesseract, EasyOCR, PaddleOCR).
• Familiarity with model optimization tools — TensorRT, ONNX, or OpenVINO for faster inference. • Exposure to edge AI deployment on NVIDIA Jetson or similar embedded hardware. • Knowledge of video streaming protocols (RTSP, WebRTC) and tools like FFmpeg or GStreamer. • Experience with MLOps tools such as MLflow, Weights & Biases, or DVC for experiment tracking. • Understanding of multi-object tracking algorithms (DeepSORT, ByteTrack, etc.). • Contributions to open-source CV projects or research publications are a plus |
| · Languages: Python (primary), Bash
· Vision Libraries: OpenCV, torchvision, Pillow, scikit-image, Albumentations · Deep Learning: PyTorch, TensorFlow, Keras · Detection Models: YOLOv5, YOLOv8, Faster R-CNN, SSD, EfficientDet · OCR Tools: Tesseract, EasyOCR, PaddleOCR · Model Optimization: TensorRT, ONNX, OpenVINO, ONNX Runtime · MLOps & Tracking: MLflow, Weights & Biases, DVC, Label Studio · Cloud & DevOps: AWS / GCP / Azure, Docker, Kubernetes, GitHub Actions |
