Research Engineer Graduate (AI Training Systems & RL Infrastructure - Seed Infra) - 2026 Start (PhD)

About the team The Seed Infrastructures team oversees the distributed training, reinforcement learning framework, high-performance inference, and heterogeneous hardware compilation technologies for AI foundation models.

We are looking for talented individuals to join our team. As a graduate, you will get opportunities to pursue bold ideas, tackle complex challenges, and unlock limitless growth. Successful candidates must be able to commit to an onboarding date by the end of the year. Please state your availability and graduation date clearly in your resume.

Responsibilities

• Conduct research and development on large-scale AI infrastructure to support efficient training and post-training of foundation models, multimodal LLMs, and image/video generation models.
• Design and optimize distributed training strategies, including data/model/tensor/pipeline/expert parallelism, computation-communication overlap, and large-scale GPU cluster scaling.
• Prototype and improve end-to-end reinforcement learning (RL) training systems, covering rollout generation, policy optimization, evaluation, and iterative deployment workflows.
• Build scalable and fault-tolerant infrastructure that operates reliably under dynamic workloads and heterogeneous compute environments.
• Analyze performance bottlenecks across the training stack (e.g., networking, scheduling, GPU memory management), and develop principled optimization approaches to improve throughput, efficiency, and stability.
• Develop tooling, monitoring, debugging, and observability frameworks to ensure reliability of large-scale training and RL systems.
• Collaborate with researchers and engineers on system-algorithm co-design, translating research prototypes into scalable, production-ready infrastructure systems.Minimum Qualifications
• Individuals who are completing or have recently completed a PhD degree in Computer Science, Electrical Engineering, Electrical and Computer Engineering, Physics, Mathematics, or a related discipline.
• Strong background in distributed systems, large-scale machine learning systems, or deep learning infrastructure.
• Research or hands-on experience in training or optimizing large-scale models (e.g., LLMs, multimodal models, RL systems).
• Understanding of parallelism strategies (e.g., data, model/tensor, pipeline, expert parallelism) and distributed training concepts.
• Familiarity with reinforcement learning workflows such as rollout generation, policy optimization, and evaluation loops.
• Proficiency in programming (e.g., Python and/or C++) and experience with modern ML frameworks (e.g., PyTorch and distributed training tools).