模型训练-大模型安全群组
Core
Develop and maintain LLM safety capabilities through prompt engineering, agent construction, automated evaluation operations, and security testing to ensure content compliance and model robustness.
Role type
Senior LLM Safety Engineer (Evaluation & Prompt Engineering)
Builds
Automated evaluation suites, safety testing frameworks, and secure LLM products
Domain
Artificial Intelligence / Large Language Models / Content Safety
Deliverable
production ML models
Required skills
Python, SQL, Prompt Engineering, LLM Evaluation, Automated Scripting, Data Analysis, Security Testing, Cross-functional Collaboration
Preferred skills
Experience with AI/ML platforms, Knowledge of small languages, Model training background
Technologies
Python, SQL, LLM Evaluation Platforms, Agent Frameworks
Responsibilities
Write prompts and build agents to enhance model safety; Operate automated evaluation systems and generate assessment reports; Explore novel prompt attack vectors and build comprehensive security testing systems; Conduct security inspections and risk closure for live products; Analyze data to optimize delivery processes and quality.
Seniority
Senior, hands-on IC
