Senior Edge AI Engineer

Focus: NVIDIA Jetson · Computer Vision · VLMs & LLMs Engagement: Part-Time / Contract | Location: Remote About the Role Shyld AI is a Silicon Valley startup developing advanced AI-powered systems for real-world environments, including hospitals. We are looking for a Senior Edge AI Engineer with strong hands-on experience deploying, optimizing, and running AI models on resource-constrained edge devices, particularly NVIDIA Jetson platforms. The ideal candidate has extensive experience with computer vision, open-source Vision-Language Models (VLMs), and Large Language Models (LLMs), with a proven track record of delivering production-level solutions. Must-Have Skills - Edge AI Deployment: Strong experience deploying deep learning models on edge hardware with real-world constraints on latency, memory, and compute. - NVIDIA Jetson: Hands-on experience with Jetson Orin Nano, Orin NX, or AGX Orin, including performance optimization and production deployment. - Computer Vision: Strong experience with object detection, segmentation, tracking, and visual scene understanding. - Model Optimization: Practical experience with quantization (INT8/INT4), pruning, distillation, and inference acceleration using TensorRT, ONNX Runtime, CUDA, or similar technologies. - Model Fine-Tuning: Experience fine-tuning and adapting pretrained AI models, including LoRA/QLoRA and domain-specific training. - Open-Source VLMs & LLMs: Hands-on experience deploying and optimizing models such as Qwen, Llama, InternVL, or similar open-source architectures. - Python & PyTorch: Strong software engineering skills, including production-quality Python development and deep learning frameworks. - Production Experience: Proven industry experience delivering AI solutions deployed on real hardware, beyond academic research, simulations, or small proof-of-concept projects. Nice-to-Have Skills - Real-time inference and multimodal AI pipelines. - NVIDIA JetPack, CUDA, TensorRT-LLM, llama.cpp, or vLLM. - RGB-D cameras, visual localization, and robotic perception. - ROS2 integration and edge robotics systems. - Experience with model benchmarking, profiling, and hardware-aware architecture selection. What You'll Work On - Deploy and optimize computer vision models and VLMs on NVIDIA Jetson devices. - Fine-tune models for specialized visual understanding tasks. - Reduce inference latency, memory consumption, and computational requirements. - Build efficient pipelines for processing camera streams and multimodal sensor data. - Evaluate open-source AI models and optimize them for reliable, real-world operation. How to Apply Please share: - Your experience with NVIDIA Jetson and deployed edge AI systems. - Specific production projects involving model deployment, fine-tuning, or quantization. - Your experience with open-source VLMs and LLMs. - GitHub repositories, technical portfolios, or examples of relevant deployed solutions. • Working Hours: Consistent availability between 9:00 AM to 5:00 PM Pacific Time (PT), • Strong, clear technical communication during Pacific business hours.