AI Infrastructure Engineer Graduate (Live Recommendation Architecture)

About the role

TikTok's Live Recommendation Architecture team builds and optimizes the architecture of its live broadcast recommendation system, focusing on system stability, high availability, online services and offline data flow performance. It works closely with applied machine learning engineers to build scalable systems supporting new algorithms. The company is looking for a graduate to join the team, starting in 2027.

Key responsibilities

• Optimize model performance and memory efficiency on GPU-based systems for TikTok Live recommendation scenarios
• Deploy and scale high-throughput training and inference pipelines for live broadcast recommendation models
• Develop tools, libraries and compiler optimizations to accelerate deep learning workloads for live streaming
• Analyze system performance through GPU profiling, kernel analysis and throughput tuning for real-time workloads
• Build and maintain high-performance computing frameworks and storage systems for model training and inference

Required profile

• Completing or recently completed a Bachelor's or Master's degree in AI, software development, computer science, computer engineering or a related field
• Experience programming in at least one of C, C++, Java or Golang
• Effective communication skills, a sense of ownership and drive
• Commitment to an onboarding date by the end of the year; applications reviewed on a rolling basis

Required skills

• Solid programming skills in C++, CUDA, Triton and Python
• Familiarity with GPU architecture and distributed training is highly desirable
• Preferred: production-grade training and inference systems, hands-on model optimization, distributed training frameworks (NCCL, Horovod, DeepSpeed, FSDP), deep learning compilers (TVM, LLVM), livestream or real-time systems experience, open-source contributions