Agent数据/评测专家 - AI数据与安全
Core
Designing data production and evaluation strategies for voice AI Agents, building datasets, and establishing automated evaluation pipelines to ensure model accuracy and performance.
Role type
Senior IC AI Data & Evaluation Engineer (Voice Agents)
Builds
Evaluation datasets, automated assessment pipelines, and data quality assets for voice Agent models.
Domain
AI / Large Language Models / Voice Agents / Data Engineering
Deliverable
production ML models
Required skills
LLM-as-Judge, Agent-as-Judge, Python, dataset construction, rubric design, model problem attribution, automated evaluation pipeline engineering, multi-turn interaction analysis, tool use understanding, trajectory analysis
Preferred skills
Knowledge engineering, multimodal data, agentic AI trends, AI development tools (TRAE/Claude Code/Cursor)
Responsibilities
Designing data and evaluation schemes for voice Agents, building datasets and executing evaluations, maintaining automated evaluation systems with Python, collaborating on data standard iteration and expert training, analyzing model performance and attributing issues
Seniority
Senior, hands-on IC
