Senior SRE Engineer (HPC & Cloud)

TL;DR

Senior SRE Engineer (HPC & Cloud): Managing large-scale Linux environments and HPC clusters with an accent on automation, cloud infrastructure, and storage optimization. Focus on building internal AI platforms, optimizing CI/CD pipelines, and ensuring the reliability of high-performance compute services.

Location: Taiwan

Company

hirify.global is a quantitative trading firm specializing in high-frequency trading and advanced research.

What you will do

• Manage large-scale Linux environments, focusing on troubleshooting and deep root-cause analysis.
• Develop maintainable automation using Bash, Ansible, and Python for infrastructure operations.
• Operate HPC clusters (Slurm) and maintain high-performance storage solutions like Lustre and NAS.
• Manage multi-cloud infrastructure across AWS, GCP, and Alibaba Cloud using Terraform and AWS CDK.
• Build and operate Docker/Kubernetes (ECS, EKS) environments and design GitLab CI/CD pipelines.
• Develop internal AI platforms, chatbots, and agents utilizing LangChain, Bedrock, and Elasticsearch RAG.

Requirements

• 5+ years of hands-on Linux systems administration and infrastructure operations experience.
• Deep knowledge of Linux internals including process, memory, filesystem, networking, and cgroups.
• Proficiency in Bash/Shell scripting and Python for data processing and API services.
• Solid experience with RAID, filesystem selection, and shared storage operations (NFS/SMB).
• Experience with public cloud providers (AWS/GCP/Alibaba) and IaC tooling (Terraform/Ansible).
• Ability to drive complex technical subsystems end-to-end with strong autonomy and minimal supervision.

Nice to have

• Experience with HPC schedulers (Slurm, PBS, LSF) or parallel filesystems (Lustre, GPFS).
• Advanced Linux performance analysis skills using eBPF, perf, or ftrace.
• Database operations experience with MySQL or ClickHouse.
• GPU server operations including NVIDIA driver management, CUDA toolkit, and Slurm GRES configuration.
• LLM application development experience with LangChain or RAG.