Research Engineer – Agent Systems & AI Coding Environment (Seed Infra Platform)

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.

Responsibilities

• Build agent harness and execution environments for AI coding and knowledge tasks (code execution, tool integration, sandboxing, system interaction)
• Develop scalable orchestration frameworks for multi-step agent workflows (planning, tool use, memory, coordination)
• Design evaluation and benchmarking systems to measure agent performance across complex, long-horizon tasks
• Improve agent performance via prompting, data curation, and post-training in collaboration with model and RL teams
• Partner with research and product teams to productionize agent systems

Minimum Qualifications

• Bachelor’s degree or above in Computer Science or a related field
• Strong programming skills (Python or similar) and solid system-building experience
• Experience with LLM-based systems, pipelines, or tool-integrated workflows
• Understanding of system design and building scalable infrastructure
• Experience building agent harness / runtime systems or AI coding environments
• Experience with Docker, Kubernetes or similar orchestration systems, distributed job execution, and containerized/sandboxed code execution environments

Preferred Qualifications

• Familiarity with evaluation frameworks, prompting / finetuning / RL, or large-scale experimentation