Staff Software Engineer, On-Device Machine Learning - Google
ML Engineer
We're looking for a ML Engineer focused on advancing capabilities through data analysis and experimentation. This lead role requires 8+ years of relevant experience.
About the role
Minimum qualifications:
Bachelor’s degree or equivalent practical experience.
8 years of experience in software development.
5 years of experience with software development in C++ programming language.
5 years of experience testing, and launching software products, and 3 years of experience with software design and architecture.
5 years of experience leading ML design and optimizing end-to-end Machine Learning (e.g., model deployment, model evaluation, data processing, debugging, fine tuning).
Preferred qualifications:
Master’s degree or PhD in Engineering, Computer Science, or a related technical field.
8 years of experience with data structures and algorithms.
3 years of experience in a technical leadership role leading project teams and setting technical direction.
3 years of experience working in a complex, matrixed organization involving cross-functional, or cross-business projects.
Experience in deploying on-device Machine Learning models.
Responsibilities
Enable one or more areas in Google AI Edge Developer Tools such as AI Edge Gallery (Gemma's showcase app for on-device Gemma 0-day), Developer Device Platform (Device farm with AI, enabling developers to test/eval/deploy apps and models across device), and AI Edge Eval (innovative eval framework for on-device model and harness).
Train task-specific models of in-app AI enabling Gemma model/runtime co-design (quantization, conversion), build industry-leading on-device AI solutions (voice translate, generative image editing)
Enable tooling of GTM for on-device Gemma 0-day (AI Edge Gallery), helping developers to test/eval/deploy across devices.
Enable on-device deployment of key models across various accelerators (GPU/Pixel TPU/NPUs/CPU) on Android, Chrome, and more.
Improve performance of on-device model inference via optimizations in the model presentation, on-device runtime and kernel implementation.