Agent工程师实习生（J107185）
Core
Build and optimize intelligent evaluation workflows for A/B experiments using Large Language Model (LLM) Agents to estimate pre-launch effects and automate reporting.
Role type
Intern, LLM Agent Engineer (Evaluation & Experimentation)
Builds
Automated evaluation pipelines and data strategies for A/B testing
Domain
AI / Machine Learning / Experimentation
Deliverable
production ML models
Required skills
Prompt Engineering, LLM Agent implementation, Workflow orchestration, Python, SQL, Data analysis, A/B testing principles
Preferred skills
Data mining, Data processing, Content creation, LLM data annotation
Technologies
Python, SQL, LLMs
Responsibilities
Estimate pre-launch effects and simulate user evaluations for A/B experiments; Design and iterate evaluation strategies to align simulated and real experiment conclusions; Collaborate with product and algorithm teams on scenario definition and data standardization.
