Agent策略算法实习生（J98721）
Core
Research and develop core strategies for Large Language Model (LLM) agents, focusing on training, evaluation, and deployment of creative agents for complex content production workflows.
Role type
Intern, LLM Agent Strategy Researcher
Builds
Creative agents with content-aware capabilities and multi-task scheduling for content generation
Domain
Artificial Intelligence / Large Language Models / Agent Systems
Deliverable
production ML models
Required skills
Python, PyTorch or PaddlePaddle, Transformer architecture, LLM fundamentals, Prompt Engineering, Reinforcement Learning (PPO/DPO), Agent frameworks (LangChain/Dify)
Preferred skills
Long-context RAG, Multimodal understanding and alignment, DeepResearch frameworks, Multi-agent collaboration systems
Responsibilities
Develop and iterate core agent strategies; Participate in model training and evaluation; Build creative agents for content production; Reproduce and validate frontier agent technologies in business scenarios
