ByteDance
Student Researcher - (Seed Model - Infra - Compiler) - 2026 Start (PhD)
San Jose, California · Internship · Posted Oct 28, 2026
- Experience
- No specific number of years stated.
- Education
- Currently pursuing a PhD in Computer Science, Electrical Engineering, or a related technical field.
About the job
Contribute to AI compiler and runtime improvements that optimize foundation-model training and inference across GPU and NPU hardware.
What you'll do
- Develop MLIR-based compiler passes for graph lowering, optimization, and code generation.
- Optimize model execution and support deployment pipelines across hardware backends.
- Benchmark and profile large-scale models; assist with distributed training and inference acceleration.
What they ask for
- Currently pursuing a PhD in Computer Science, Electrical Engineering, or a related technical field.
- Experience using or developing open-source LLM inference frameworks such as vLLM or SGLang, and proficiency in at least one deep learning framework with model inference workflow experience.
- Understanding of modern computing systems and familiarity with compilers, model optimization pipelines, or related execution workflows.
- Able to commit to 12 weeks of work in 2026.
Nice to have
- Experience with distributed or large-scale ML systems and related optimizations.
- Experience with GPU/TPU/NPU programming, performance optimization, or high-performance computing and communication.
Tools
MLIR, vLLM, SGLang, PyTorch, Megatron, DeepSpeed, JAX, PyTorch Dynamo
AI focus
AI compiler optimization and acceleration of foundation-model training and inference.
Model-written summary · the original posting is authoritative