大模型算法工程师（J104493）
Core
Design and develop multi-modal agent systems with autonomous planning, complex reasoning, and dynamic decision-making capabilities for real-world business scenarios.
Role type
Senior IC large model algorithm engineer (agents)
Builds
Multi-modal agent architectures with autonomous planning and tool invocation
Domain
Artificial Intelligence / Large Language Models / Agent Systems
Deliverable
production ML models
Required skills
Agent system design, ReAct, CoT, Plan-and-Execute, PyTorch, LangChain, LLM pre-training, SFT, RLHF, Reinforcement Learning, RAG, Multi-agent collaboration
Preferred skills
Top-tier conference publications (CVPR/ACL/ICML), ACM/IOI/IMO competition awards, Core contributions to major open-source projects
Responsibilities
Lead core algorithm R&D and iteration for agent modules like intent understanding and memory management; Build multi-dimensional evaluation systems for effect, efficiency, and cost; Drive model capability boundaries using frontier technologies like RAG and CoT.
