ByteDance
Software Engineer Graduate (Data-Speech-Product RD-Engineering-US) - 2027 Start
San Jose, California · Full-time · Posted Oct 28, 2026
- Education
- Completing or recently completed a Bachelor's/Master's degree in computer engineering or a related discipline.
- Visa sponsorship
- Candidates must establish authorization to work in the United States. The Company does not provide sponsorship or immigration-related benefits.
About the job
Design and implement techniques to optimize AI models and their training and inference for efficient, large-scale production deployment.
What you'll do
- Develop model optimization algorithms, including quantization, pruning, knowledge distillation, and efficient architectures.
- Build benchmarking frameworks and optimize training and inference pipelines on GPUs and distributed systems.
- Work with ML researchers to bring optimized models into production and stay current with research in model efficiency, compilers, and systems.
What they ask for
- Completing or recently completed a Bachelor's/Master's degree in computer engineering or a related discipline.
- Strong coding skills in Python and C++.
Nice to have
- Experience with deep learning frameworks and distributed training systems.
- Understanding of computer architecture, parallel computing, and GPU acceleration; familiarity with CUDA, Triton, or ML compilers such as TVM, XLA, or TensorRT is a plus.
Tools
Python, C++, CUDA, Triton, TVM, XLA, TensorRT
AI focus
AI model optimization, including quantization, pruning, knowledge distillation, efficient architectures, and optimization of training and inference.
Model-written summary · the original posting is authoritative