Key responsibilities
1. Responsible for overall technical architecture design of the company's AI Agent system, including core components such as perception-planning-execution loops, multi-agent collaboration mechanisms, tool invocation frameworks, and long-term memory modules.
2. Design and implement key infrastructure for Agents, including but not limited to: task planning engines, tool invocation and security control platforms, memory systems (short-term/long-term), Agent collaboration workflows (based on frameworks like LangGraph), and real-time environment perception interfaces.
3. Lead technology stack selection and integration for AI Agents, including Agent frameworks like LangChain/LangGraph, RAG-enhanced retrieval, multimodal interaction modules, and build high-availability Agent service platforms.
4. Optimize AI Agent system performance and reliability, focusing on ensuring task decomposition and execution accuracy, tool invocation security (e.g., sandbox mechanisms, permission control), and multi-Agent collaboration efficiency.
5. Track cutting-edge Agent technologies (such as multimodal Agents, autonomous evolution mechanisms, reinforcement learning applications), drive technology adoption in business scenarios, and lead technology roadmap development.
6. Guide R&D teams through the full process from Agent scenario design, tool integration, memory module development to system deployment, ensuring architecture capabilities are iterative and scalable.