Google seeks a mid-level Software Engineer in Zürich to optimize ML models for DSP and HTP architectures, ensuring high performance within strict power and thermal constraints.
This position is part of the Platforms and Devices team, which integrates Google AI, software, and hardware across computing platforms and first-party devices. The engineer will tackle critical projects with the flexibility to shift focus as business needs evolve, aiming to define efficiency standards at the intersection of machine learning and system architecture.
Responsibilities
- Leverage Digital Signal Processors (DSP) and Hexagon Tensor Processors (HTP) to speed up machine learning models while keeping thermal impact low.
- Perform detailed power profiling and apply hardware-specific tweaks to ensure consistent feature performance across all scenarios.
- Manage the full lifecycle of performance optimization to keep models operating within a defined power budget.
- Establish efficiency benchmarks for upcoming hardware by bridging machine learning algorithms with system design.
Requirements
- Hold a Bachelor’s degree or possess equivalent practical experience.
- Have at least two years of software development experience using C++ or Python, or one year if holding an advanced degree.
- Demonstrate one year of experience with machine learning infrastructure, including model deployment, evaluation, data processing, and debugging.
Nice to have
- Master’s degree or PhD in Computer Science or a related technical discipline.
- Background in creating accessible technologies.
- Strong grasp of machine learning principles to guarantee algorithms run within tight power limits.
- Up-to-date knowledge of recent advances in DSP, HTP, and system-on-chip (SoC) architectures.
About the company
The company focuses on researching and designing new technologies to make user interactions with computing devices faster and more seamless. Engineers are encouraged to be versatile, show leadership, and tackle complex problems across the entire technology stack.
- Teams work on innovations that streamline how users interact with computing devices.
- Staff are expected to demonstrate versatility and leadership while solving full-stack challenges.
Quelle: öffentlich zugängliche Karriereseite des Arbeitgebers. Batchly ist nicht der Arbeitgeber und steht nicht notwendigerweise in einem Vertragsverhältnis mit dem Unternehmen.