- Shanghai, China
- On-site
- Full-time · Principal
- 3+ years
Principal Engineer System Application Engineering role focusing on AI model development/localization and designing heterogeneous TinyML architecture for E84.
Responsibilities
- Align priority and collaborate with PL & AE team on AI model development/localization
- Design heterogeneous TinyML architecture for E84 to split work load between NNLite wake-up domain and Ethos‑U55 inference domain
- Firmware development on ModusToolbox: M55 for pre/post-processing and Ethos‑U55 inference; M33 for NNLite wake-up
- Benchmark Ethos‑U55 MAC utilization, latency, TCM/SRAM peak, NNLite sleep/wake current
- Prototype emerging TinyML use cases: audio classification, IMU gesture, vibration anomaly detection, tiny VLM on ultra-low-power MCUs
Requirements
- Bachelor/Master in EE, CS or related
- Nice to have: Infineon PSoC Edge E84 / E83
- Nice to have: DeepCraft Studio NNLite kernel optimization
- Nice to have: wake-word detection (WWD) Helium
- Nice to have: MVE vector optimization