Responsibilities
- Optimize the computational performance of recommendation mid-platform models for training and inference.
- Design and develop high-performance general-purpose and business-customized kernel libraries.
- Improve computation and communication parallelism and kernel performance.
- Develop model compilation capabilities involving graph optimization, kernel fusion, computation scheduling, and code generation.
- Perform full-stack performance analysis, bottleneck diagnosis, and optimization in collaboration with business and algorithm teams.
- Build reusable performance-optimization components, toolchains, and platform capabilities for internal businesses.
Requirements
- Be completing or have recently completed a bachelor’s or master’s degree in software development, computer science, computer engineering, or a related technical discipline.
- Have development and optimization experience with mainstream model compilation stacks such as TVM, MLIR, or XLA.
- Be proficient in C/C++ and familiar with assembly, CPU/GPU architecture, cache mechanisms, and high-performance kernel development.
- Understand the underlying principles of deep learning frameworks, computational graphs, and inference/training execution processes.
- Have experience implementing model graph optimization and compilation optimization.
- Demonstrate independent problem-solving, performance troubleshooting, and practical optimization abilities.
- Preferred: experience with hardware-software co-design and heterogeneous computing projects.
- Preferred: contributions to or development of open-source deep learning kernel libraries, compilers, or inference engines.
- Preferred: in-depth research experience with the architecture and mechanisms of TensorFlow, PyTorch, MxNet, or another machine learning framework.
Benefits
- Graduate role with opportunities to pursue challenging ideas and complex machine learning infrastructure problems.
- Candidates must be able to commit to an onboarding date by the end of the year and should state availability and graduation date in their resume.
- Applications are reviewed on a rolling basis; candidates may apply to a maximum of two company or affiliate positions globally.
Categories
About ByteDance
ByteDance is a global incubator of platforms at the cutting edge of commerce, content, entertainment and enterprise services - over 2.5bn people interact with ByteDance products including TikTok. Creation is the core of ByteDance's purpose. Our products are built to help imaginations thrive. This is doubly true of the teams that make our innovations possible. Together, we inspire creativity and enrich life - a mission we aim towards achieving every day. At ByteDance, we create together and grow together. That's how we drive impact - for ourselves, our company, and the users we serve. We are committed to building a safe, healthy and positive online environment for all our users. We have over 110,000 employees based in more than 30 countries globally. Join us.