LLM-Powered AI Infrastructure Engineer

TikTok; Singapore; On-site; SGD 60,000 - 120,000; Full time

Job summary

TikTok in Singapore seeks graduates for a role in E-commerce Recommendation Infra, working on scalable systems for LLM-based recommendations. You will integrate LLM tech, deploy training/inference optimizations, and collaborate with ML engineers to push end-to-end generative recommendation tech.

The team focuses on lifelong behavior modeling, model scale-up, and deploying foundation models with open-source tools. Apply early and include graduation date in your resume.

Qualifications

• Bachelor's or Master's in AI, CS, Software or related field.
• Strong coding skills in C++, CUDA, Triton and Python.
• Familiarity with GPU architecture and distributed training is desirable.
• Proficiency in C, C++, Java or Golang and good communication.
• Ownership mindset and ability to work with teams.

Responsibilities

• Deeply integrate LLM and recommendation technologies and optimize training/inference.
• Develop automated model optimization for PyTorch-based frameworks.
• Design and implement next-gen model infra with open-source components.
• Research, innovate, and productionize end-to-end generative recommendations.
• Optimize throughput for high-performance GPU kernels and multi-GPU setups.

Skills

C++/CUDA/Triton/Python

GPU architecture

Distributed training

C/Java/Golang

Communication and ownership

Education

Bachelor's or Master's in AI/CS/Software

Tools

Megatron-LM

DeepSpeed

vLLM

SGLang

TensorRT-LLM

Job description

TikTok in Singapore seeks graduates for a role in E-commerce Recommendation Infra, working on scalable systems for LLM-based recommendations. You will integrate LLM tech, deploy training/inference optimizations, and collaborate with ML engineers to push end-to-end generative recommendation tech.

The team focuses on lifelong behavior modeling, model scale-up, and deploying foundation models with open-source tools. Apply early and include graduation date in your resume.