大模型测试工程师（北京/深圳）
Core
Designing and executing quality assurance for large language models (LLMs), covering RL+agent workflows, inference performance, and protocol compatibility.
Role type
LLM testing engineer
Builds
Production-ready LLM services and inference pipelines
Domain
Artificial Intelligence / Large Language Models
Deliverable
production ML models
Required skills
LLM testing execution, RL+agent workflow validation, inference performance testing, protocol compatibility testing, black-box testing, performance testing, stability testing, LLM evaluation metrics, test case design, service architecture understanding
Preferred skills
Experience with LLM deployment pipelines, familiarity with mainstream interface protocols, test automation
Technologies
RL, agent, LLM, inference engines, mainstream interface protocols
Responsibilities
Understand LLM business requirements and scope boundaries, design test cases for normal/abnormal/boundary/high-concurrency scenarios, execute RL+agent flow and performance tests, coordinate with model evaluation teams for data and standard alignment, optimize test case libraries and automation processes
Seniority
Mid-level, hands-on IC
