Graduate AI Model Optimization Engineer

ByteDance; San Jose (CA); On-site; USD 128,000 - 256,000; Full time

Job summary

ByteDance is seeking an AI Model Optimization Engineer to design and implement techniques to speed up AI models, improve efficiency, and simplify deployment at scale. You will collaborate with research and engineering to push performance in production.

Responsibilities include developing optimization algorithms, benchmarking, and optimizing GPU pipelines across distributed systems. A strong background in Python/C++, DL frameworks, and ML compilers is expected.

Qualifications

• Bachelor's or master's in computer engineering or related field.
• Strong coding skills in Python and C++.
• Experience with deep learning frameworks and distributed training systems.
• Familiarity with GPU programming (CUDA, Triton, or similar) and ML compilers like TVM/XLA/TensorRT is a plus.

Responsibilities

• Develop and implement model optimization algorithms (quantization, pruning, distillation, efficient architectures).
• Build and maintain performance benchmarking frameworks for large-scale training and inference.
• Optimize training and inference pipelines on GPUs and across distributed systems.
• Collaborate with ML researchers to productionize optimized models.
• Stay current with research in model efficiency, compilers, and systems.

Skills

Python

C++

GPU programming (CUDA)

Deep learning frameworks

Distributed training systems

ML compilers familiarity

Education

Bachelor's or Master's in Computer Engineering or related

Tools

TVM

XLA

TensorRT

CUDA

Job description

ByteDance is seeking an AI Model Optimization Engineer to design and implement techniques to speed up AI models, improve efficiency, and simplify deployment at scale. You will collaborate with research and engineering to push performance in production.

Responsibilities include developing optimization algorithms, benchmarking, and optimizing GPU pipelines across distributed systems. A strong background in Python/C++, DL frameworks, and ML compilers is expected.