AI Infrastructure Optimization Engineer Intern (TikTok Global E-Commerce Recommendation & Search Architecture) - 2027 Start
Core
Building and optimizing scalable infrastructure for LLM-based recommendation systems in e-commerce live-streaming and short video scenarios.
Role type
Intern, AI Infrastructure Optimization Engineer
Builds
Distributed inference serving architectures and model optimization toolchains for large-scale LLM deployment
Domain
E-commerce, Generative AI, Large Language Models
Deliverable
production ML models
Required skills
C++, CUDA, Python, GPU architecture, distributed training, deep learning frameworks (PyTorch, TensorFlow), LLM fundamentals
Preferred skills
LLM inference acceleration (quantization, KV cache optimization, FlashAttention), LLM systems (Megatron-LM, DeepSpeed, vLLM, TensorRT-LLM), recommendation systems
Technologies
PyTorch, TensorFlow, Megatron-LM, DeepSpeed, vLLM, TensorRT-LLM
Responsibilities
Research and implement LLM-based recommendation foundation model training and inference optimization; Build model optimization toolchains including quantization, pruning, and distillation; Design and implement distributed inference serving architectures
Seniority
Intern