Research Engineer — Training Performance & ML Compilation (Torch Compile) - Seed Infra

ByteDance; Seattle (WA); On-site; USD 180,000 - 240,000; Full time

Job summary

ByteDance Seed team in Seattle is hiring for a role focused on optimizing training performance for large-scale foundation models. You will push compiler-level improvements, extend ML compilation capabilities with PyTorch stack, and design high-performance GPU kernels for training workloads.

Collaborate with research and infrastructure teams to profile, diagnose, and resolve bottlenecks. Ideal candidates have CUDA/Triton experience and strong C++/Python skills.

Qualifications

• Bachelor's degree in Computer Science, Electrical Engineering, or related field.
• Strong proficiency in C/C++ and Python; solid data structures and systems programming.
• Hands-on training-side performance optimization for DL workloads.
• Experience writing and optimizing GPU kernels (CUDA, Triton).
• Experience with the PyTorch compilation stack (Inductor/FX) or Triton kernel development.

Responsibilities

• Optimize training performance for large-scale foundation models through compiler-level techniques.
• Develop ML compilation capabilities based on PyTorch stack to improve training efficiency on GPUs.
• Design and optimize high-performance GPU kernels for training workloads.
• Profile and analyze large-scale training jobs to identify bottlenecks with research and infra teams.

Skills

C/C++

Python

Performance optimization

GPU kernels

PyTorch compilation stack

Education

Bachelor's degree in Computer Science or Electrical Engineering

Tools

CUDA

Triton

PyTorch

Job description

Join us as we work together to inspire creativity and enrich life around the globe.

Location:

Seattle

Team:

Technology

Employment Type:

Regular

Job Code:

A105162A

Share this listing:

Responsibilities

• Optimize training performance for large-scale foundation models through compiler-level techniques, including graph optimization, operator fusion, and kernel generation.
• Develop and extend ML compilation capabilities based on the PyTorch compilation stack (e.g. FX, Dynamo, Inductor) to improve training efficiency across heterogeneous GPU platforms.
• Design and optimize high-performance GPU kernels for training workloads.
• Conduct performance profiling and analysis of large-scale training jobs; identify and resolve bottlenecks in collaboration with research and infrastructure teams.

Qualifications

Minimum Qualification(s)

• Bachelor's degree or above in Computer Science, Electrical Engineering, or a related field.
• Strong proficiency in C/C++ and Python; solid foundations in algorithms, data structures, and systems programming.
• Hands-on experience in training-side performance optimization for deep learning workloads.
• Hands-on experience writing and optimizing GPU kernels (e.g., CUDA, Triton).
• Experience with the PyTorch compilation stack, meeting at least one of the following: direct experience using, debugging, or extending Inductor or FX; proficiency in Triton kernel development; or solid experience with PyTorch computation graph work (graph optimization, graph capture, operator fusion).

Preferred Qualification(s)

• Experience with TorchDynamo or bytecode-level program transformation.
• Experience with Triton compiler internals or other ML compiler backends (e.g., MLIR, LLVM).
• Contributions to related open-source projects (e.g., PyTorch, Triton, FlashAttention).
• Publications in relevant venues (e.g., MLSys, OSDI, ASPLOS).

Job Information

About Doubao (Seed)

Established in 2023, the ByteDance Seed team is dedicated to pioneering new paths toward artificial general intelligence. We aspire to advance the frontier of intelligence to drive progress for both technology and society.

With a long-term vision for the AI sector, the Seed team's research spans MLLM, GenMedia, AI for Science, and Robotics. We maintain a global presence with laboratories and career opportunities across China, Singapore, and the United States. To date, we have launched industry-leading general foundation models and cutting-edge multimodal capabilities. Our technology powers over 50 application scenarios — including Doubao, Jimeng, TRAE, Dola and Dreamnia — and serves enterprise customers through Volcano Engine and BytePlus. Third-party data shows that the Doubao App ranks first in user volume in the Chinese market, while Doubao foundation models lead the industry in average daily token consumption.

Why Join ByteDance

Inspiring creativity is at the core of ByteDance's mission. Our innovative products are built to help people authentically express themselves, discover and connect – and our global, diverse teams make that possible. Together, we create value for our communities, inspire creativity and enrich life - a mission we work towards every day.

As ByteDancers, we strive to do great things with great people. We lead with curiosity, humility, and a desire to make impact in a rapidly growing tech company. By constantly iterating and fostering an "Always Day 1" mindset, we achieve meaningful breakthroughs for ourselves, our Company, and our users. When we create and grow together, the possibilities are limitless. Join us.

Diversity & Inclusion

ByteDance is committed to creating an inclusive space where employees are valued for their skills, experiences, and unique perspectives. Our platform connects people from across the globe and so does our workplace. At ByteDance, our mission is to inspire creativity and enrich life. To achieve that goal, we are committed to celebrating our diverse voices and to creating an environment that reflects the many communities we reach. We are passionate about this and hope you are too.

Reasonable Accommodation

ByteDance is committed to providing reasonable accommodations in our recruitment processes for candidates with disabilities, pregnancy, sincerely held religious beliefs or other reasons protected by applicable laws. If you need assistance or a reasonable accommodation, please reach out to us at https://tinyurl.com/RA-request