腾讯会议-评测产品经理
Core
Design and build evaluation systems for ASR and large model products, covering general and specialized capabilities like speech recognition accuracy, conversational understanding, and multi-modal tasks.
Role type
Product Manager (Evaluation & Benchmarking)
Builds
Evaluation frameworks, benchmark datasets, and analysis tools for AI products
Domain
Artificial Intelligence / Large Language Models / Speech Recognition
Deliverable
production ML models
Required skills
ASR and large model evaluation, benchmark research, data analysis, badcase mining, metric definition, cross-functional collaboration, technical documentation
Preferred skills
Experience with AI search or voice assistants, evaluation platform construction, English technical reading
Technologies
ASR, Large Language Models, WER/CER metrics, multi-modal systems
Responsibilities
Define evaluation dimensions and metrics for product capabilities; Research and implement industry benchmarks for business scenarios; Analyze model performance to identify optimization opportunities; Monitor user feedback to establish a closed-loop improvement mechanism; Collaborate with engineering and algorithm teams to integrate evaluation into the product lifecycle.
