Edge AI Embedded Firmware Engineer (Low-Power)

Sonova; Stäfa; Hybrid; CHF 90.000 - 130.000; Vollzeit

Benefits dieser Stelle

Flexible hybrid working model

Training opportunities

Growth opportunities

Zusammenfassung

Sonova in Switzerland is seeking an Embedded Software Engineer to join the R&D team. You will develop the embedded software stack for next-generation hearing devices, focusing on neural networks on a deeply embedded platform and low-power constraints.

You will design, debug on target hardware, deploy NN models, collaborate with AI and DSP experts, and ensure high-quality software through reviews, testing and documentation, contributing to edge AI capabilities for hearing instruments.

Qualifikationen

• Degree in CS, EE, or related technical field.
• Hands-on embedded software development experience on resource-constrained systems.
• Familiarity with deploying neural networks on embedded platforms.

Aufgaben

• Design, develop and maintain firmware for a deeply embedded platform operating under strict latency, memory, compute, and energy constraints.
• Debug complex issues directly on target hardware and simulators.
• Contribute to the integration and deployment of neural networks in hearing-device firmware with AI/DSP experts.
• Advise AI developers on tailoring neural networks to the embedded platform’s constraints.
• Collaborate with engineering teams to develop prototypes and translate them into production-ready features.
• Maintain high software quality through clean-code practices, reviews, testing, documentation, and root-cause analysis.

Kenntnisse

Embedded C/C++

OO Design

Python

Performance profiling

Embedded tooling

English fluency

Neural networks on embedded

Ausbildung

Degree in CS/EE

Tools

TensorFlow Lite

ONNX

TVM

ExecuTorch

Jobbeschreibung

Sonova in Switzerland is seeking an Embedded Software Engineer to join the R&D team. You will develop the embedded software stack for next-generation hearing devices, focusing on neural networks on a deeply embedded platform and low-power constraints.

You will design, debug on target hardware, deploy NN models, collaborate with AI and DSP experts, and ensure high-quality software through reviews, testing and documentation, contributing to edge AI capabilities for hearing instruments.