About this role
Project description Designing, optimizing, and productizing high-performance computer vision and AI pipelines for consumer device - with focus on real-time performance, power efficiency, and production robustness across embedded platforms.
Responsibilities
- Design and implement advanced computer vision and image processing pipelines optimized for real-time consumer devices.
- Collaborate with ISP, sensor, and tuning teams to optimize image quality for downstream AI and UX performance.
- Develop and deploy ML models for visual recognition, enhancement, tracking, or scene understanding.
- Optimize ML models for edge deployment (quantization, pruning, distillation, hardware-aware tuning).
- Implement performance-critical algorithms in modern C++ for embedded platforms.
- Optimize for latency, power consumption, memory footprint, and thermal constraints.
- Integrate inference engines (TFLite, TensorRT, ONNX Runtime, etc.) on target SoCs.
- Work closely with Android/Linux platform teams to integrate camera and AI pipelines.
- Define and track KPIs: FPS, power usage, memory, startup time, and accuracy.
- Profile and optimize performance across CPU/GPU/NPU/DSP.
- Drive debugging of complex system-level issues in production builds.
- Ensure robust unit testing and contribute to automated validation pipelines.
- Mentor engineers and review architecture/design proposals.
- Support product bring-up and mass production readiness.
Skills
- Must have
- Bachelor's or Master's degree in Computer Science