基座模型安全产品-AI创新业务
Core
Define safety standards, construct safe training data, and build evaluation frameworks to ensure content safety and compliance for large language model (LLM) foundations like Doubao.
Role type
Senior Product Manager (LLM Safety & Risk Governance)
Builds
LLM foundation safety training pipelines, risk governance systems, and safety evaluation frameworks
Domain
Artificial Intelligence / Large Language Models / Content Safety
Deliverable
production ML models
Required skills
LLM safety strategy definition, safe data construction, evaluation framework design, cross-functional collaboration, risk analysis, product roadmap planning
Preferred skills
Understanding of pre-training, fine-tuning, PE, RAG, and Agent architectures, user-centric business thinking, strategic foresight in AI safety
Technologies
LLMs, RAG, PE, Agent
Responsibilities
Define safety standards and data norms for LLM training; Build and optimize LLM safety evaluation systems; Collaborate with algorithm and engineering teams to implement safety strategies.
Seniority
Senior, hands-on IC with team leadership
