Principal Architect
Core
Design the software backbone for a unified, high-performance video analytics engine running on embedded hardware, bridging systems engineering and AI research to enable real-time AI inference on edge devices.
Role type
Principal Architect (Embedded AI & High-Performance Computing)
Builds
A modular, graph-based execution engine for video analytics and AI models on diverse edge hardware platforms.
Domain
Embedded AI, Edge Computing, Computer Vision, High-Performance Systems
Deliverable
production ML models
Required skills
Modern C++ (14/17/20), System Architecture, Heterogeneous Resource Orchestration, AI Model Optimization (Quantization/Pruning), Memory & Data Flow Design, Hardware-agnostic Abstraction Layer design
Preferred skills
Experience with TensorRT, SNPE/QNN, OpenVINO, GStreamer, MediaPipe, DeepStream, SoC architectures (Qualcomm, Ambarella, NVIDIA), System Profiling (Perf, eBPF, Nsight Systems)
Technologies
C++, TensorRT, SNPE, QNN, OpenVINO, GStreamer, MediaPipe, DeepStream, eBPF, Nsight Systems
Responsibilities
Architect a modular, high-throughput Video & AI Pipeline Engine with dynamic graph construction; Design scheduling logic to balance workloads across CPU, GPU, DSP, and NPU; Define architectural support for Model Quantization and Network Pruning; Architect zero-copy data transport mechanisms; Create a hardware-agnostic Abstraction Layer for scaling from low-power cameras to high-performance processors.
Seniority
Principal, hands-on IC with strategy & mentorship
