AI Compute Intern - Network Site Reliability Engineer

About the role

Tencent Overseas IT is looking for a motivated Network Site Reliability Engineer (SRE) intern for its AI Compute team. The intern supports the daily operations and maintenance of AI-accelerated high-performance computing (HPC) infrastructure, working closely with internal business teams and vendor engineering teams to build and operate clusters. The AI Compute Centre sits within Overseas IT and covers the full lifecycle of AI infrastructure clusters, from requirement gathering and capacity planning through design, delivery, operations and DevOps.

Key responsibilities

• Support the deployment, configuration and maintenance of high-end servers, storage servers, networking equipment and software in secure environments.
• Assist with hardware diagnostics, system checks and firmware updates.
• Help deliver tailored customer environments such as bare-metal systems, HPC clusters, Kubernetes and Slurm.
• Provide first-line engineering support for onsite issues, including hardware, network and software troubleshooting and firmware compliance.
• Document incidents, resolutions and lessons learned, and keep documentation accurate and up to date.
• Take part in team meetings and knowledge-sharing sessions.
• Check cluster network connectivity (InfiniBand or RoCE links), run tests and flag anomalies.
• Monitor dashboards for compute, storage and management networks and record bandwidth, latency and RDMA results against baselines.
• Support incident investigations by collecting switch, NIC and host logs and tracking vendor follow-up.

Required profile

• Currently pursuing or recently completed a Bachelor's or Master's degree in computer engineering, computer science or a related field.
• Ability to work independently and as part of a team.

Required skills

• Basic understanding of server hardware, firmware lifecycle and Linux environments, and awareness of physical and system-level security standards.
• Exposure to Bash or Python scripting.
• Familiarity with configuration management, CI/CD tools, workload managers and cluster software (Slurm, Kubernetes, BCM), and observability tools (Prometheus, Grafana, ELK).