Job Description

Your Role:

1. Deploy and optimize AI models on edge devices such as mobile, embedded, and IoT platforms.
2. Efficiently port and extend open-source inference frameworks like ONNXRuntime, TFLite, etc.
3. Build a unified deployment toolchain to enable the principle of "develop once, deploy anywhere".
4. Apply lightweight optimization techniques such as pruning and quantization to improve model performance.
5. Establish performance evaluation systems to monitor inference speed, resource usage, and key efficiency metrics.
6. Write standardized deployment documentation and build reusable component libraries.

✅ Required Qualifications

1. Master’s degree or above in Computer Science, Electrical Engineering, Automation, or a related field
2. 5+ years of experience in edge computing, embedded systems, or AI deployment
3. Strong proficiency in C++17 with solid system-level programming skills
4. Familiar with at least two inference engines such as TFLite, MNN, ONNXRuntime, or NCNN
5. Hands-on experience with model conversion tools (e.g., tf-lite-converter, ONNX export pipelines)
6. Proficient in Linux development environments, including CMake and cross-compilation workflows
7. Practical experience deploying models on embedded devices such as Raspberry Pi or NVIDIA Jetson

✨ Preferred Qualifications

1. Experience with model quantization (INT8/FP16) and distillation
2. Familiarity with edge hardware acceleration (e.g., NPU, GPU)
3. Rust development experience
4. Understanding of edge-cloud collaborative deployment architectures
5. Experience optimizing for ARM architecture, including NEON instruction set

Location:

Silicon Valley (preferred), Suzhou, or Remote

Job Tags

Similar Jobs