混元Agent数据与环境平台工程师（深圳/北京/上海）
Core
Building and platformizing intelligent agent interaction environments for large language model training and evaluation, managing datasets across pretrain/posttrain/eval stages, and connecting data, task frameworks, sandbox environments, and evaluation results.
Role type
Senior IC backend engineer (LLM agent platform & data infrastructure)
Builds
Agent interaction environments, dataset management platforms, and task execution frameworks for LLM training and evaluation
Domain
Artificial Intelligence / Large Language Models / Data Infrastructure
Deliverable
production ML models
Required skills
Python, Docker, Kubernetes, backend service development, data modeling, system architecture design, data versioning, quality validation, automation pipeline design
Preferred skills
Experience with self-developed, third-party, and open-source environment integration, complex system task handling, multi-turn user environment support
Technologies
Python, Docker, Kubernetes
Responsibilities
Design and implement interfaces and data standards for agent interaction environments; Build and manage dataset capabilities including metadata, content, versions, and quality status; Collaborate on dataset and framework design for complex and long-task scenarios; Establish traceability and reproducibility across the data-to-evaluation pipeline.
