模型评测产品经理 - AI创新业务
Core
Build and manage the evaluation system for AI core experiences in the Maoxiang app, focusing on role-playing dialogue, voice, images, multi-branch storylines, and memory systems.
Role type
Product Manager (AI Evaluation & Metrics)
Builds
Automated evaluation pipelines, benchmarks, and datasets for multimodal AI models
Domain
Consumer AI / AIGC / Interactive Narrative
Deliverable
production ML models
Required skills
Product management, data analysis, experimental design, SQL, Python, multimodal evaluation, A/B testing
Preferred skills
Experience in content quality, recommendation strategies, experience metrics, conversational AI, AIGC, role-playing, interactive storytelling
Technologies
SQL, Python
Responsibilities
Design A/B experiments to assess model iteration impact on user retention and engagement; Collaborate with algorithm and data teams to drive model optimization based on evaluation results; Establish automated evaluation mechanisms combining automated testing and human annotation.
