大模型内容安全算法工程师-集团信息系统
Core
Design and implement algorithms for privacy compliance and data security in AI applications, specifically focusing on sensitive information identification and content risk assessment within LLM interaction chains.
Role type
Senior IC large model content safety algorithm engineer
Builds
Sensitive discovery and classification systems, risk governance frameworks, and privacy defense mechanisms for LLMs
Domain
AI / Large Language Models / Content Safety / Data Privacy
Deliverable
production ML models
Required skills
NLP algorithms (text classification, intent understanding, knowledge graphs), multimodal understanding, deep learning frameworks (TensorFlow, PyTorch), Python/Go/C++, graph algorithms, transfer learning
Preferred skills
AI Agent applications, semantic sensitive identification, context leakage detection, weak supervision, synthetic data, adversarial robustness, automated red/blue teaming
Responsibilities
Analyze large-scale structured and unstructured data to build risk identification capabilities; collaborate with product and engineering teams to design interception, desensitization, and audit strategies; explore frontier technologies to enhance solution competitiveness
Seniority
Senior, hands-on IC
