About the position
ByteDance, founded in 2012, is on a mission to inspire creativity and enrich life through its diverse suite of products, including TikTok and various platforms tailored for the Chinese market. The company emphasizes the importance of creation, not just in its products but also within its teams, fostering an environment where challenges are seen as opportunities for learning and innovation. The ByteDance Doubao (Seed) Team, established in 2023, is focused on pioneering advanced AI foundation models, with research areas that include deep learning, reinforcement learning, and AI safety. The team operates globally, with labs in China, Singapore, and the US, and is committed to leveraging substantial data and computing resources to drive technological advancements. The Machine Learning (ML) System sub-team is dedicated to developing and maintaining distributed ML training and inference systems, providing high-performance and scalable solutions for LLM/AIGC/AGI.
Responsibilities
- Developing and optimizing LLM inference framework.
- Performing GPU and CUDA performance optimization to create a high-performance LLM inference engine.
Requirements
- Strong experience in machine learning and system engineering.
- Proficiency in GPU programming and CUDA optimization.
- Experience with distributed systems and high-performance computing.
Nice-to-haves
- Familiarity with AI foundation models and multimodal capabilities.
- Experience in working with large-scale heterogeneous systems.
