Skymizer is developing a software stack for deploying AI models at various types of AI system-on-chips, from high-performance cloud to cost-effective edge devices.

As a software engineer of AI calibration, you will help develop a so-called “calibrator” tool for quantizing the data precision of AI models under the target machine’s constraints.

Your contribution is important for enabling those exciting deep neural network models really working in the daily life. Without calibrator, a deep neural network model trained under floating-point precision cannot run on a hardware with only small bit-width (such as int8 precision), which benefits cost and efficiency.

This job is also challenging, though. You will develop algorithms to minimize the quantization error.

We are heavily using C++. You should be familar with modern C++ techniques, and be able to program in a logical, reusable, and timely fashion.

Responsibility

• Propose quantization strategies that can minimize the precision loss.
• Deliver high-quality implementation on schedule.

Minimum qualifications

• Experience with C++ programming.
• Experience with neumerical analysis.
• Bachelor’s or above degree in Computer Science, Mathematic Science or related technical discipline.

Preferred qualifications

• Knowledge of computer architecture.
• Knowledge of deep neural network model.
• Knowledge of model compression and quantization.
• Knowledge of compiler framework such as MLIR, LLVM, or GCC.