• Design and develop computer vision algorithms for real-time inference on edge devices. 
• Optimize deep learning models for performance and memory on platforms like NVIDIA Jetson (Nano, Xavier, Orin). 
• Deploy, benchmark, and fine-tune models using frameworks such as TensorRT, ONNX, and DeepStream. 
• Work closely with hardware and embedded teams to ensure seamless integration and performance. 
• Evaluate and experiment with Vision-Language Models (VLMs) for potential on-device applications. 
• Continuously research the latest advancements in edge AI and computer vision.

• Bachelor's or Master’s degree in Data science, Computer Science or a related field. 
• 2+ years of experience in computer vision and deep learning (PyTorch, TensorFlow, OpenCV, etc.). 
• Proven experience fine-tuning and deploying models for face recognition, object detection/tracking, license plate recognition, and event detection. 
• Experience with deploying AI models on edge devices (Jetson Nano/Xavier/Orin, Raspberry Pi, etc.). 
• Skilled software developer with experience in C/C++, Python, and/or CUDA, experience with edge device deployment is preferred (TensorRT, TFlite, ONNX, etc) 
• Familiarity with tools like OpenCV, GStreamer, or DeepStream SDK. 

Nice to Have: 

• Experience working with Vision-Language Models (VLMs) like Intern, Qwen, Llama 
• Contributions to open-source CV/AI projects or research publications.