AI Research Engineer - Model Acceleration & Quantization

intel; Prescott (AZ); On-site; USD 52,000 - 82,000; Full time

Job summary

Intel Labs China is seeking a PhD-level researcher to perform cutting-edge AI research and engineering in Bejing. You will collaborate with a world-class team to develop efficient algorithms that compress and accelerate large AI models for deployment across client, edge, cloud, and data center platforms.

The role emphasizes post-training quantization techniques, several AI inference frameworks, and strong software skills in PyTorch, C/C++, Python, and shell scripting.

Qualifications

• Ph.D. graduate or equivalent in AI, CS, EE, Automation, or related field.
• Strong expertise in developing AI algorithms for compressing and accelerating large models.
• Experience with quantization of large models (4-bit, 1.58-bit) desirable.

Responsibilities

• Conduct cutting-edge AI research and engineering at Intel Labs China.
• Collaborate with team members to develop efficient algorithmic solutions for accelerating large AI models.
• Focus on deploying models across Intel computing platforms for client, edge, cloud, and data center applications.

Skills

AI research

Model compression

Post-training quantization

PyTorch

C/C++

Python

Shell scripting

OpenVINO

Education

Ph.D. in AI/CS/EE/Automation

Tools

vLLM

Ollama

OpenVINO

Job description

Intel Labs China is seeking a PhD-level researcher to perform cutting-edge AI research and engineering in Bejing. You will collaborate with a world-class team to develop efficient algorithms that compress and accelerate large AI models for deployment across client, edge, cloud, and data center platforms.

The role emphasizes post-training quantization techniques, several AI inference frameworks, and strong software skills in PyTorch, C/C++, Python, and shell scripting.