Software Engineer, AI Infrastructure - USDS

About the Team

The USDS JV Reliability and Efficiency team works with engineering and product teams to build and run large-scale, globally distributed, observable, fault-tolerant systems for TikTok state-of-the-Art Recommendation and Search engines. Engineers on this team will deliver on production ownership and be responsible for observability and automation across complex, large-scale service mesh architectures.

Responsibilities

• Assist in the deployment, operation, and optimization of Recommendation and Search systems throughout their entire lifecycle.
• Develop and maintain internal software and automation scripts to eliminate manual tasks and improve overall R&D velocity.
• Support the availability of large-scale services deployed across global data centers through proactive monitoring and maintenance.
• Track and report on cloud resource utilization to help the team meet efficiency goals and maintain cluster SLAs.
• Configure and manage monitoring stacks to track service latency, health metrics, and system performance.

Minimum Qualifications

• Bachelor's degree in Computer Science or a related technical field, with 1+ years of professional experience in software development or infrastructure.
• Foundational knowledge of distributed systems and a strong interest in troubleshooting complex technical issues.
• Effective communication skills and a proactive approach to solving engineering bottlenecks.
• Experience with agentic frameworks and tools development.

Preferred Qualifications

• Familiarity with container orchestration tools like Kubernetes.
• Basic understanding of the machine learning lifecycle or model inference workflows.