ByteDance
Senior Research Scientist - Machine Learning System
San Jose, California · Senior · Full-time · Posted Oct 28, 2026
- Education
- Bachelor's degree or above in computer/electronics/automation/software, etc.
About the job
Develop and optimize LLM inference frameworks and GPU/CUDA performance to build a high-performance LLM inference engine.
What you'll do
- Develop and optimize LLM inference frameworks.
- Optimize GPU and CUDA performance for LLM inference.
What they ask for
- Bachelor’s degree or above in computer, electronics, automation, software, or a related major.
- Proficiency in C/C++, algorithms, and data structures; familiarity with Python.
- Understanding of deep learning principles, neural network architectures, and deep learning training frameworks such as Pytorch.
Nice to have
- GPU high-performance computing optimization with CUDA, computer architecture, parallel computing, memory access optimization, or low-bit computing.
- Familiarity with TensorRT-LLM, ORCA, or VLLM; knowledge of LLMs or experience accelerating and optimizing LLM models.
Tools
C/C++, Python, Pytorch, CUDA, TensorRT-LLM, ORCA, VLLM
AI focus
Develop and optimize LLM inference systems and accelerate LLM models.
Model-written summary · the original posting is authoritative