LLM大模型评测产品经理-豆包手机助手
Core
Define product experience and evaluation standards for Large Language Models (LLMs) from a user perspective, building assessment systems for real-world scenarios to drive model and product iteration.
Role type
LLM Product Manager (Evaluation & Strategy)
Builds
LLM-powered mobile assistant features (Doubao)
Domain
AI / Large Language Models / Mobile Applications
Deliverable
production ML models
Required skills
LLM evaluation, multi-modal assessment, Agent product evaluation, structured analysis, experimental design, data-driven problem solving, cross-functional collaboration
Preferred skills
Experience with LLM/multi-modal/Agent product evaluation, understanding of AI technology trends, ability to translate research into product strategy
Technologies
LLMs, Multi-modal models, Agents
Responsibilities
Define ideal model performance states and evaluation metrics; Build evaluation frameworks for real-world application scenarios; Collaborate with R&D and data science to identify model defects and optimization opportunities; Drive cross-team execution to validate and optimize model behaviors and user experience; Iterate evaluation methodologies based on industry research and business scenarios.
