LLM-Driven AI Infrastructure Intern

TikTok; Singapore; On-site; SGD 13,000 - 20,000; Full time

Job summary

TikTok is seeking an AI Infrastructure Optimization Engineer Intern for a 2027 start in Singapore. You will research and implement optimization techniques for large-scale LLM-based e-commerce recommendations, and help build end-to-end inference pipelines.

You will work with C++/CUDA/Python, PyTorch/TensorFlow, and distributed systems, gaining hands-on experience with Megatron-LM, DeepSpeed, and other LLM tools while contributing to real-world deployment.

Qualifications

• Pursuing a Bachelor’s or Master’s degree in CS/CE/IS.
• Strong programming in C++/CUDA/Python; basic GPU knowledge.
• Familiarity with PyTorch and TensorFlow; knowledge of LLMs.

Responsibilities

• Research and implement LLM-based optimization for recommendation use cases.
• Build model optimization toolchains including quantization, pruning, distillation.
• Design distributed inference serving architectures for large-scale LLMs.

Skills

C++

CUDA

Python

GPU architecture

Distributed training

PyTorch

TensorFlow

LLM acceleration

Education

Bachelor's degree in CS/CE/IS

Master's degree in CS/CE/IS

Tools

Megatron-LM

DeepSpeed

vLLM

TensorRT-LLM

Job description

TikTok is seeking an AI Infrastructure Optimization Engineer Intern for a 2027 start in Singapore. You will research and implement optimization techniques for large-scale LLM-based e-commerce recommendations, and help build end-to-end inference pipelines.

You will work with C++/CUDA/Python, PyTorch/TensorFlow, and distributed systems, gaining hands-on experience with Megatron-LM, DeepSpeed, and other LLM tools while contributing to real-world deployment.