About the Role

The Computer Vision Engineer will be responsible for developing advanced deep learning solutions and real-time inference systems that power high-performance visual intelligence applications. This role involves building end-to-end computer vision pipelines—from dataset preparation and model development to deployment in production environments. You will work extensively with PyTorch and modern vision frameworks to design, train, optimize, and deploy models capable of handling complex tasks such as detection, segmentation, classification, OCR, and tracking.

You will be expected to work with state-of-the-art architectures, optimize models for GPU and edge environments, and ensure production readiness through scalable training pipelines and robust deployment workflows. This position requires strong technical depth, hands-on model development experience, and the ability to deliver practical, high-accuracy solutions tailored for real-world use cases.

Key Responsibilities

• Build deep learning models for detection, segmentation, classification, OCR, and tracking.

• Use PyTorch, TorchVision, OpenCV, and SOTA frameworks (MMDetection, Ultralytics).

• Implement modern architectures (YOLOv8/11, ViT, SAM, DETR variants).

• Optimize models for GPU and edge devices (ONNX, TensorRT, quantization).

• Build scalable training pipelines with reproducible experiment tracking. Deploy models through FastAPI / Docker with real-time performance constraints.

Required Skills

• Solid proficiency in Python + PyTorch.

• Hands-on experience training CNN/Transformer vision models.

• Strong foundation in CV fundamentals (augmentation, metrics, losses).

• Experience with GPU training environments (CUDA basics).

Bonus Skills

• Jetson/edge AI optimization.

• MLflow / W&B tracking.

Experience with multimodal CV or diffusion-based models.

If you would like to apply for this position, send your CV to [email protected]