Position Overview We are seeking an experienced Infrastructure Engineer to architect and manage our AI computing infrastructure. The ideal candidate will have extensive experience in building and scaling ML infrastructure, with particular emphasis on distributed training systems and GPU cluster management. Key Responsibilities Design and implement high-performance computing infrastructure for large-scale AI model training Manage and optimize GPU clusters for distributed training workloads Build and maintain container orchestration systems for ML workflows Implement efficient resource allocation and scheduling systems Design and maintain monitoring and alerting systems for compute infrastructure Optimize infrastructure costs while maintaining performance Collaborate with ML teams to support their computing needs Ensure system reliability, security, and scalability Required Qualifications Master's degree in Computer Science, Systems Engineering, or related field 8+ years of experience in infrastructure engineering, with focus on ML/AI infrastructure Strong experience with: GPU cluster management and optimization Kubernetes and container orchestration Linux system administration Infrastructure as Code (IaC) Proven track record in building large-scale computing systems Experience with major cloud providers (AWS/GCP/Azure or Alibaba Cloud/Tencent Cloud etc) Preferred Qualifications Experience with ML infrastructure at major tech companies Knowledge of distributed training systems (PyTorch DDP, Horovod) Familiarity with ML frameworks and their infrastructure requirements Experience with high-performance networking (InfiniBand, RDMA) Background in performance optimization and troubleshooting Understanding of ML workload characteristics Bilingual proficiency (English/Chinese) Technical Skills Computing Infrastructure GPU Clusters: NVIDIA DGX, GPU management tools Distributed Systems: Slurm, Kubernetes ML Platforms: Kubeflow, Ray Job Scheduling: YARN, Slurm Cloud & Networking Cloud Platforms: International: AWS, GCP, Azure China: Alibaba Cloud, Tencent Cloud Networking: InfiniBand, RDMA, TCP/IP optimization Load Balancing: HAProxy, NGINX Infrastructure Management Container Technologies: Docker, Kubernetes, Singularity IaC: Terraform, Ansible, CloudFormation CI/CD: Jenkins, GitLab CI Monitoring: Prometheus, Grafana, ELK Stack Development Languages: Python, Go, Shell scripting Version Control: Git Documentation: Markdown, Confluence What We Offer Opportunity to build cutting-edge AI infrastructure Competitive salary and equity package Access to latest hardware and technologies Professional development opportunities Comprehensive health benefits Learning and conference budget Location Hong Kong (on-site, Hong Kong Science and Technology Park) Expected Impact Design and implement next-generation AI computing infrastructure Optimize resource utilization and cost efficiency Improve training speed and efficiency for AI models Build scalable and reliable systems Projects You'll Work On Building automated GPU cluster management systems Implementing efficient resource scheduling for ML workloads Optimizing distributed training infrastructure Setting up monitoring and observability systems Designing disaster recovery and backup solutions