About the role#
The Seed LLM Post Training team researches technologies for unified multimodal large models. This internship focuses on post-training phases, including SFT, RM, RL, and self-learning. You will work on optimizing model capabilities in reasoning, coding, and agent behavior.
What you'll need#
- Currently pursuing a PhD in Computer Science, AI, or a related field.
- Research experience in reinforcement learning, sequential decision-making, or agent behavior.
- First-author publications in conferences like NeurIPS, ICLR, or ICML.
- Proficiency in programming and experimentation with RL or LLM frameworks.
- Experience with LLM agents, tool use, or prompt-based control is preferred.
- Familiarity with environments such as WebArena, ALFWorld, or programmatic reasoning tasks.
- Knowledge of RL techniques like reward shaping, memory augmentation, or curriculum learning.
- You must obtain and maintain work authorization in the United States for the duration of the internship.
What you'll do#
- Explore large-scale models and optimize systems.
- Construct data and perform instruction tuning and preference alignment.
- Improve model performance in areas like reasoning, math, and code.
- Conduct research to explore future use cases for these models.
Location & details#
- Location: San Jose, California.
- Type: Full-time internship.
- Modality: On-site.
- Process: Applications are reviewed on a rolling basis.
- Note: Sponsorship is not provided for this position.


