Embedded Software Engineer, Machine Learning Runtime, Silicon

Google is seeking an Embedded Software Engineer to join their team working on the Google Tensor, a custom System-on-Chip (SoC) that powers the latest Pixel phones. The role focuses on developing the mobile software stack for the Tensor SoC, with an emphasis on deploying on-device ML frameworks for various user experiences including Camera, Speech, Generative AI, and more.

As an Embedded Software Engineer, you will be responsible for:

1. Developing on-device Runtime software stack to deploy ML models on mobile devices.
2. Tracking and driving performance and power optimizations for on-device models.
3. Supporting customers with deploying their ML models to the Tensor TPU.
4. Collaborating with application software teams to prototype and enable new use cases on Tensor SoCs.
5. Triaging product or system issues and debugging/tracking/resolving by analyzing the sources of issues and their impact on software, hardware, and quality.

The ideal candidate should have a strong background in embedded operating systems, software development (particularly in C++), and experience with software frameworks for mobile ML or graphics (e.g., TFLite, Vulkan, OpenCL, CUDA). Knowledge of machine learning algorithms, computer architecture, and on-device machine learning (including NLP, image processing, and on-device Generative AI) is highly valued.

Google offers a competitive salary range of $136,000-$200,000 + bonus + equity + benefits for this role. The company is committed to creating a diverse and inclusive work environment, providing equal opportunities to all qualified candidates.

Join Google's mission to organize the world's information and make it universally accessible and useful by working on cutting-edge technologies that combine the best of Google AI, Software, and Hardware to create radically helpful experiences.