DuMate策略PM（Agent评测方向）（J104381）
Core
Build and execute an evaluation system for long-horizon AI Agents, defining testing standards, cases, and processes to quantify strategy performance and drive product iteration.
Role type
Strategy Product Manager (Agent Evaluation)
Builds
Evaluation frameworks, testing suites, and product capabilities for long-horizon AI Agents
Domain
Artificial Intelligence / Large Language Models / Agent Systems
Deliverable
production ML models
Required skills
Strategy formulation, requirement decomposition, data-driven optimization, logical reasoning, problem analysis, cross-functional collaboration
Preferred skills
Computer Science/AI/Software Engineering/Data Science background, hands-on experience with AI/Agents/LLMs, understanding of long-horizon task pain points
Technologies
AI frameworks, Agent platforms, LLMs
Responsibilities
Design long-horizon task evaluation standards and test cases; Validate and review core skills and kits for product launch; Iterate evaluation schemes and metrics based on business scenarios; Establish SOPs for the evaluation platform to manage annotation tasks
Seniority
Mid-level, hands-on IC
