Yuezhou Liu

Shanghai, China · lewis-remove-.liu@outlook.com

Senior Cloud and AI Infrastructure Engineer with 17+ years of experience spanning telecom core networks, cloud platforms (Kubernetes, OpenStack), and GPU/AI infrastructure. Proven track record in building and operating large-scale distributed systems, leading technical teams, and landing AI workloads on cloud infrastructure. Currently focused on SAP AI Core deployment on Alibaba Cloud. Deep expertise in Kubernetes, bare-metal provisioning (Metal3), GPU cluster networking (InfiniBand/RDMA), and LLM deployment.

Skills

Cloud & Container Orchestration
  • Kubernetes
  • OpenStack
  • Docker
  • Metal3
  • Helm
  • Rook-Ceph
AI / GPU Infrastructure
  • Nvidia BCM
  • InfiniBand
  • GPFS
  • RDMA
  • GPU Operator
  • vLLM
Programming & Automation
  • Shell/Bash
  • Python
  • Ansible
  • Erlang
  • C++
  • Java
Telecom & Networking
  • 3GPP (2G-5G)
  • Core Network (SGSN/MME)
  • IP Networking
  • OVS/DPDK/SR-IOV
  • DNS
  • Load Balancing
DevOps & Platform
  • CI/CD
  • Linux
  • Git
  • Prometheus
  • Grafana
  • Alibaba Cloud

Creations

Full-Page Screenshot (Claude Code Skill)

A zero-dependency Claude Code skill that captures full-page screenshots using Chrome DevTools Protocol (CDP). Handles long pages with automatic scrolling and stitching, no Puppeteer or Playwright required.

Read more..

AI Proxy for SAP AI Core

A multi-provider AI proxy gateway designed for SAP AI Core, supporting OpenAI, Anthropic Claude, Google Gemini, and LiteLLM endpoints with automatic deployment routing. Enables seamless AI model integration within SAP's enterprise cloud infrastructure on Alibaba Cloud.

Read more..

Remote Exec — SSH Reverse Command Execution

A tool that establishes a persistent SSH connection from Machine A to Machine B, enabling remote command execution on Machine A from Machine B. Useful for managing machines behind NAT or firewalls.

Read more..

Wait! There's more..

See all Creations for more examples!

Experience

Senior DevOps Engineer

China Datacom (SAP Cloud China)

SAP AI Core China landing on Alibaba Cloud.

  • Responsible for infrastructure deployment and operations of SAP’s AI platform on the Chinese public cloud environment

September 2025 - Present

Principal DevOps Engineer

Yoocar

Built and operated AI GPU cluster infrastructure from the ground up.

  • Designed and deployed H20 GPU server clusters managed by Nvidia BCM (BaseCommand Manager)
  • Implemented high-performance networking with InfiniBand, RDMA, and IBM GPFS high-speed storage
  • Built Kubernetes platform on bare-metal with automated provisioning based on Metal3/ClusterAPI
  • Deployed and fine-tuned Large Language Models (DeepSeek, LLaMA) using vLLM, NIM, and Ollama
  • Participated in drafting National Standards for Intelligent Computing Terms

April 2023 - July 2025

Senior Cloud Support Engineer / Team Leader

Ericsson

Led cloud infrastructure team for Ericsson’s telecom cloud products (CCD & CEE).

  • Technical lead for CCD (CNCF-certified Kubernetes platform) and CEE (Mirantis OpenStack platform)
  • Kubernetes lifecycle management automation development (Heat/Ansible)
  • OpenStack components (Nova/Neutron) customization for China Mobile (Python/Ansible/Shell)
  • Prototype research on bare-metal K8S deployment based on Metal3
  • Supported China Mobile 5GC CP1 performance testing, ranking #1 among all vendors
  • Customized CEE (OpenStack) to support 1,500 physical nodes for China Mobile
  • NFV virtual layer networking (OVS/DPDK/SR-IOV) design, development, and automation
  • Supported KDDI Japan CEE deployment

September 2016 - March 2023

Senior Support Engineer

Ericsson

Expert-level 2G/3G/4G/5G core network technical support.

  • Analyzed core network signaling and helped operators configure and troubleshoot SGSN/MME
  • 7x24 R&D Level 3 emergency support for global telecom operators
  • Deployed and tested virtual core network (vEPC) on Ericsson OpenStack cloud platform
  • On-site support for 2016 G20 Summit (Zhejiang Telecom core network)
  • China Mobile NB-IoT test specification drafting
  • Softbank VoLTE pre-launch functional testing

July 2012 - August 2016

Software Developer

Tellabs

Developed Tellabs 6300 Network Management System (NMS).

  • Module design and development using C++, Shell, and TCL

October 2011 - July 2012

Software Developer

HP (Hewlett-Packard)

Developed HP OpenView TeMIP (Telecom Management Information Platform).

  • Server alarm module development (Shell/C++)
  • Windows client GUI development (MFC)
  • Web client development (Java)

February 2008 - September 2011

Education

JiangSu University

Bachelor of Engineering
Information Security

2004 - 2008
Nifty tech tag lists from Wouter Beeftink