SRE Intern: Build & Optimize AI Compute Clusters

Tencent; Singapore; Vor Ort; SGD 20.000 - 27.000; Teilzeit

Zusammenfassung

Tencent seeks a motivated Site Reliability Engineer (SRE) Intern to join the AI Compute team, supporting daily operations and maintenance of AI-accelerated HPC infrastructure. You will work closely with internal business teams and vendor engineers to build and operate clusters and gain exposure to cutting-edge AI infrastructure.

The role involves collaborating on bare-metal systems, HPC clusters, Kubernetes, and Slurm, plus monitoring via Prometheus, Grafana, and ELK.

Qualifikationen

• Pursuing or recently completed Bachelor's or Master's in computer engineering, CS, or related field.
• Familiarity with Linux and server hardware, security basics.
• Exposure to scripting and cluster software is a plus.

Aufgaben

• Support deployment, configuration, and maintenance of HPC infrastructure.
• Assist with hardware diagnostics and firmware updates.
• Collaborate with teams to deliver bare-metal, HPC, Kubernetes, and Slurm environments.
• Provide first-line engineering support for onsite issues and troubleshooting.
• Document incident details, resolutions and lessons learned.
• Maintain up-to-date documentation for knowledge sharing.
• Participate in team meetings and knowledge-sharing sessions.

Kenntnisse

Bash/Python scripting

Linux environments

Configuration management

CI/CD tools

Cluster software (Slurm/Kubernetes)

Observability tools (Prometheus/Graf</

Networking basics (InfiniBand/RoCE)

Documentation

Ausbildung

Bachelor's or Master's degree in computer engineering or computer science

Tools

NVIDIA BCM

Jobbeschreibung

Tencent seeks a motivated Site Reliability Engineer (SRE) Intern to join the AI Compute team, supporting daily operations and maintenance of AI-accelerated HPC infrastructure. You will work closely with internal business teams and vendor engineers to build and operate clusters and gain exposure to cutting-edge AI infrastructure.

The role involves collaborating on bare-metal systems, HPC clusters, Kubernetes, and Slurm, plus monitoring via Prometheus, Grafana, and ELK.