Architect, design, and build distributed, asynchronous AI systems that seamlessly span wearable smart glasses, mobile devices, and cloud infrastructure, managing real-time multimodal streaming, state management, buffering, backpressure, and fault tolerance. Lead the end-to-end integration of computer vision, language, and multimodal models into production systems, designing routing and decision-making mechanisms grounded in model uncertainty, confidence calibration, abstention, and verification.
Drive system-wide performance and reliability across device, mobile, and cloud boundaries, making architectural trade-offs across latency, bandwidth, compute, memory, battery life, thermal limits, and inference cost. Partner with machine learning researchers and hardware teams to co-design constrained edge and mobile inference pipelines, leveraging quantization, model compression, and hardware acceleration, alongside scalable cloud AI serving and service orchestration.
Establish evaluation frameworks, production observability, and debugging tools to measure and improve end-to-end system performance and user experience.
Job description provided by Google.
View the original posting
