Site Reliability Engineer (Seed Model - Infrastructures - SRE)

ByteDance; San Jose (CA); On-site; USD 140,000 - 210,000; Full time

Job summary

ByteDance in San Jose seeks an experienced Seed Infra SRE to sustain stable ML compute infrastructure behind foundation models. You will manage large GPU clusters and distributed training platforms, ensuring reliable deployment and performance.

You’ll collaborate across regions, handle disaster recovery, capacity planning, and on-call support, with hands-on work on Kubernetes, Docker, and cloud resources to optimize utilization.

Qualifications

• 1+ years of experience in DevOps, SRE or infrastructure engineering projects.
• Proficient in at least one or two of Go / Python / Shell in a Linux environment.
• Experience operating resource management and job scheduling systems for distributed systems; familiar with the Kubernetes ecosystem and architecture.

Responsibilities

• Maintain the stable operation of machine learning systems and support the stages of large-model development, training and deployment.
• Manage and plan the company's GPU resources, including cost and budget, across GPU / CPU machines, storage and related resources; provide resource data to support management decisions.
• Drive stability governance for clusters and production services; improve resource utilization and operational efficiency through platform-based, systematic approaches.
• Operate disaster recovery, service deployment and machine lifecycle management across multiple regions and data centers; provide a stable and efficient GPU operating environment.
• Provide operational support for systems and services; participate in troubleshooting of production and system issues.

Skills

Go

Python

Shell

Tools

Kubernetes

Docker

Kata

Job description

About the Team

The Seed Infrastructures team oversees the distributed training, reinforcement learning framework, high-performance inference, and heterogeneous hardware compilation technologies for AI foundation models.

The Seed Infra SRE team is responsible for the delivery and stability of the compute infrastructure behind Seed's foundation models. We own the reliability of large-scale heterogeneous GPU clusters, the training and inference platforms that run on them, and the capacity, network and storage systems that keep them running. Our work directly determines how fast and how reliably we can train and serve the models.

Responsibilities

• Maintain the stable operation of machine learning systems and support the stages of large-model development, training and deployment.
• Manage and plan the company's GPU resources, including cost and budget, across GPU / CPU machines, storage and related resources; provide resource data to support management decisions.
• Drive stability governance for clusters and production services; improve resource utilization and operational efficiency through platform-based, systematic approaches.
• Operate disaster recovery, service deployment and machine lifecycle management across multiple regions and data centers; provide a stable and efficient GPU operating environment.
• Provide operational support for systems and services; participate in troubleshooting of production and system issues.

Minimum Qualifications

• 1+ years of experience in DevOps, SRE or infrastructure engineering projects.
• Proficient in at least one or two of Go / Python / Shell in a Linux environment.
• Experience operating resource management and job scheduling systems for distributed systems; familiar with the Kubernetes ecosystem and architecture.
• Familiar with containerization technologies such as Docker / Kata.
• Strong sense of ownership and responsibility, with excellent learning agility, communication, self-motivation, and teamwork skills, as well as strong analytical and logical thinking abilities to effectively abstract and break down complex business logic.

Preferred Qualifications

• Hands‑on experience operating GPU clusters or ML training / inference platforms.
• Familiarity with observability tooling (metrics, logging, alerting) and on‑call practices.
• Contributions to automation or platform tooling that measurably improved resource utilization or operational efficiency.