Edge ai principal

Overview: As an Io T and AI Architect for Edge Computing, you will be responsible for designing, developing, and deploying advanced Io T solutions with AI capabilities for edge devices. You will work on integrating AI-driven solutions directly into the edge devices to enhance real-time data processing, reduce latency, and enable efficient decision-making. You will collaborate with cross-functional teams to build scalable and resilient architectures that can handle large volumes of data generated at the edge. Key Responsibilities • Architect large-scale, distributed Edge AI systems for real-time inference across thousands of devices. • Lead hardware-aware model development optimized for deployment on platforms like NVIDIA Jetson, Qualcomm Snapdragon, Intel Movidius, and ARM Cortex-A/NPU chips. • Design pipelines for edge inferencing with low-latency, low-power, and high-accuracy constraints. • Lead performance tuning using tools like Tensor RT, ONNX Runtime, TVM, and Open VINO. • Build scalable, secure model deployment platforms using NVIDIA Triton, Edge Kubernetes (K3s), and MLOps frameworks. • Integrate AI with real-world data sources: video streams, sensor fusion (IMU, Li DAR), audio, etc. • Collaborate with firmware, hardware, and platform teams for system-level optimization. • Define strategies for OTA model updates, fleet-wide monitoring, and privacy-preserving AI. • Stay ahead of emerging trends in tiny ML, neural architecture search (NAS), zero-shot inference, and federated learning. ________________________________________ Required Technical Skills • Deep expertise in Edge AI systems, embedded deep learning, and AI model compression. • Strong knowledge of: o NVIDIA Jetson (Orin, Xavier, Nano) + Deep Stream, Tensor RT o Qualcomm QCS/QCM chips + SNPE SDK o Intel Movidius / Open VINO + Real Sense o ARM NN / Ethos-U / TFLite Micro • Model optimization: quantization (INT8/FP16), pruning, distillation, custom kernels. • Frameworks: Py Torch, Tensor Flow Lite, ONNX, TVM, Apache MXNet (optional). • Experience with edge deployment stacks: Docker, Kubernetes (K3s, Micro K8s), Yocto, RTOS. • Proficiency in Python, C++, and low-level embedded programming. • Strong understanding of network protocols, remote diagnostics, OTA updates, and device provisioning. ________________________________________ Soft Skills & Leadership • 10+ years of experience in AI/ML, with at least 3–5 years in edge/embedded environments. • Proven experience architecting and shipping commercial Edge AI products. • Strong cross-functional leadership with ability to bridge hardware, software, and product teams. • Ability to guide engineering trade-offs (latency, accuracy, memory, power) under real-world constraints. • Effective communicator with stakeholders at all levels—technical and non-technical. ________________________________________ Preferred Qualifications • MS or Ph D in Electrical Engineering, Computer Science, Robotics, or related field. • Experience with robotics (ROS/ROS2), autonomous systems, or industrial Io T deployments. • Familiarity with synthetic data generation and AI simulation environments (NVIDIA Omniverse, Unity). • Experience deploying multi-modal AI (vision, speech, sensor fusion) at the edge. • Background in cybersecurity for edge devices and AI on air-gapped networks. • Patents or publications in edge AI, embedded ML, or real-time AI inference. This role requires a blend of expertise in Io T systems, edge computing, AI, and cloud technologies, making it a critical position for organizations looking to implement smart, real-time, and scalable solutions at the edge.