Vision X Principal with Gen / AI architect.

Job Description: As an experienced professional with 15 to 23 years of expertise in Vision AI Solution, Nvidia, Computer Vision, Media, and Open Stack, you will play a crucial role in defining and leading the technical architecture for vision-based AI systems across edge and cloud environments. Your responsibilities will include designing and optimizing large-scale video analytics pipelines, architecting distributed AI systems, collaborating with cross-functional teams to develop scalable AI solutions, leading efforts in model optimization, integrating multi-modal AI, guiding platform choices, mentoring senior engineers, staying updated with emerging technologies, and contributing to internal innovation strategy. Key Responsibilities: - Define and lead the end-to-end technical architecture for vision-based AI systems across edge and cloud. - Design and optimize large-scale video analytics pipelines using NVIDIA DeepStream, TensorRT, and Triton Inference Server. - Architect distributed AI systems, including model training, deployment, inferencing, monitoring, and continuous learning. - Collaborate with product, research, and engineering teams to translate business requirements into scalable AI solutions. - Lead efforts in model optimization (quantization, pruning, distillation) for real-time performance on devices like Jetson Orin/Xavier. - Drive the integration of multi-modal AI (vision + language, 3D, audio) where applicable. - Guide platform choices (e.g., edge AI vs cloud AI trade-offs), ensuring cost-performance balance. - Mentor senior engineers and promote best practices in MLOps, system reliability, and AI observability. - Stay current with emerging technologies (e.g., NeRF, Diffusion Models, Vision Transformers, synthetic data). - Contribute to internal innovation strategy, including IP generation, publications, and external presentations. Qualification Required: - Deep expertise in computer vision, deep learning, and multi-modal AI. - Proven hands-on experience with NVIDIA Jetson, DeepStream SDK, TensorRT, Triton Inference Server, TAO Toolkit, Isaac SDK, CUDA, cuDNN, PyTorch, TensorFlow, OpenCV, GStreamer, and GPU-accelerated pipelines. - Experience deploying vision AI models at large scale (e.g., 1000+ cameras/devices or multi-GPU clusters). - Skilled in cloud-native ML infrastructure: Docker, Kubernetes, CI/CD, MLflow, Seldon, Airflow. - Proficiency in Python, C++, CUDA (or PyCUDA), and scripting. - Familiarity with 3D vision, synthetic data pipelines, and generative models (e.g., SAM, NeRF, Diffusion). - Experience in multi modal (LVM/VLM), SLMs, small LVM/VLM, Time series Gen AI models, Agentic AI, LLMOps/Edge LLMOps, Guardrails, Security in Gen AI, YOLO/Vision Transformers. (Note: Additional Details section omitted as it does not contain relevant details for the job description),