Research Scientist, LLM Pretraining; Seed-LLM

Position: Research Scientist, LLM Pretraining (Seed-LLM)
Established in 2023, the Byte Dance Seed team is dedicated to discovering new approaches to general intelligence, and pursuing the edge of intelligence. Our research spans large language models, Gen Media, AI for Science, Robotics and more. With a long-term vision and determination in AI, the Byte Dance Seed team remains committed to foundational research. We aim to become a world-class AI research team that drives real technological progress and delivers societal benefits.

With labs across China, Singapore, and the U.S., our team has already released industry-leading general-purpose large models and advanced multimodal capabilities, powering over 50 real-world applications - including Doubao, Coze, and Jimeng. The main job directions include:
1. Identify applicable ideas for optimizing large models, and apply them across models of different scales to improve model performance.
2. Explore the boundaries of ultra-large-scale models and drive extreme system-level optimizations to enhance model capability and efficiency.
3. Lead efforts in data pipeline development, instruction tuning, preference alignment, and model optimization to improve model quality and adaptability.
4. Contribute to real world applications across various domains, including creative generation, logical reasoning, code generation, and more.
5. Conduct in-depth research on future use cases of foundation models and expand the frontier of model applications.

Minimum Qualifications 1. Strong coding ability with solid knowledge of data structures and fundamental algorithms; proficient in C/C++ or Python.
2. Excellent problem-solving skills, capable of diagnosing and resolving complex challenges in large model training and deployment.
3. Strong communication and collaboration skills, with the ability to explore new technologies and drive technical advancements with the team.

Preferred Qualifications:

1. Familiarity with NLP or CV algorithms and techniques; candidates with strong research experience in related fields are preferred. Experience in large-scale model training or reinforcement learning (RL) is a plus.
2. Demonstrated leadership in high-impact projects or publications in the large-model domain is highly preferred.