Computer Vision Engineer

PranaTree is seeking a Computer Vision Engineer to develop visual intelligence solutions for real-world applications. The role focuses on training, evaluating, and deploying computer vision and multimodal AI models, preparing image and video datasets, and optimizing AI workloads for cloud, edge, and robotics environments.

Responsibilities:

• Train, fine-tune, evaluate, and deploy computer vision and multimodal AI models
• Build solutions for object detection, classification, segmentation, tracking, pose estimation, OCR, and video analytics
• Work with model families beyond YOLO, including CNN-based models, Faster R-CNN, Mask R-CNN, DETR variants, RT-DETR, Grounding DINO, SAM, CLIP, and vision-language models
• Prepare and improve image/video datasets, annotations, augmentation pipelines, and model evaluation workflows
• Optimize models for cloud, on-premises, and real-time edge deployment
• Collaborate with engineering, product, robotics, and customer teams on prototypes, pilots, and production solutions