AI Inference Infrastructure Engineer

TikTok; San Jose, Northern (CA, KY); Hybrid; USD 162,000 - 388,000; Full time

Job summary

TikTok is seeking an AI Infrastructure Engineer Graduate (Algorithm Infrastructure) in San Jose for a 2027 start. You will join a team building inference infrastructure for ultra-large-scale LMs, vision-language models, and multimodal AI systems.

The role emphasizes scalable serving, scheduling, and low-latency inference and requires a PhD in a related field. The position focuses on high-concurrency environments, kernel optimization, and production reliability across distributed systems.

Qualifications

• Candidates completing a PhD in a related technical discipline.
• Strong understanding of large-model architectures and distributed systems.
• Experience with asynchronous scheduling, load balancing, and performance optimization.

Responsibilities

• Build and evolve next-generation inference systems for large-scale online traffic and heterogeneous compute resources.
• Optimize distributed inference for 200B+ models to improve throughput and latency.
• Develop high-performance kernels for frontier model architectures (MoE, attention, multimodal fusion) using CUDA and Triton.
• Explore AI-driven infra, including AI Agents for kernel optimization and deployment pipelines.

Skills

Large-model architectures

Asynchronous scheduling

Load balancing

Performance optimization

Education

PhD in Computer Science/Software Engineering/AI

Tools

CUDA

Triton

Job description

TikTok is seeking an AI Infrastructure Engineer Graduate (Algorithm Infrastructure) in San Jose for a 2027 start. You will join a team building inference infrastructure for ultra-large-scale LMs, vision-language models, and multimodal AI systems.

The role emphasizes scalable serving, scheduling, and low-latency inference and requires a PhD in a related field. The position focuses on high-concurrency environments, kernel optimization, and production reliability across distributed systems.