AI System Researcher

Staff Researcher, AI System

Role Overview

Lenovo is seeking a Staff Researcher, AI System to research and develop advanced AI computing technologies that improve the performance and efficiency of large language models (LLMs) and AI infrastructure.

In this role, you will focus on memory optimization, AI hardware and software integration, and emerging technologies such as RISC-V. You will collaborate with engineering, hardware, and product teams to develop innovative solutions for Lenovo's AI infrastructure and intelligent computing platforms.

What You'll Do

• AI System Optimization: Design and develop large-scale memory pools and integrate them into multi-tier storage systems to improve KV Cache management and large model inference performance.

• Hardware & Software Integration: Collaborate with hardware teams to explore software-hardware co-design, including compute-storage collaboration, to improve AI system efficiency.

• RISC-V Development: Research and develop RISC-V AI technologies and explore alternative solutions for AI computing.

• Product Development: Work with business and engineering teams to integrate research outcomes into Lenovo's AI infrastructure, computing platforms, and products.

• Technology Research: Stay current with emerging AI system technologies, evaluate new developments, and share findings with the team.

Minimum Qualifications

• Master's degree or higher in Computer Science, Electronic Engineering, or a related field.

• Strong understanding of computer system architecture.

• Programming experience with Python and C/C++ in a Linux environment.

• Familiarity with I/O bus protocols such as PCIe or CXL.

• Experience with Linux driver development or related firmware development.

• Knowledge of AI hardware, including GPUs, NPUs, AI servers, and AI clusters.

• Hands-on experience with heterogeneous computing systems.

• Understanding of large language model architectures and algorithms, including models such as DeepSeek, Qwen, Kimi, and GLM.

• Familiarity with LLM inference frameworks such as vLLM, SGLang, or Llama.cpp.

• Strong analytical, problem-solving, communication, and cross-functional collaboration skills.

• Ability to learn new technologies and solve complex system-level challenges.

Preferred Qualifications

• Experience developing CXL-based memory pools.

• Experience with near-memory computing or in-memory computing projects.

• Experience with RISC-V AI technologies or ecosystem development projects.