IT之家 · ITHome
Qwen’s Jiang Guanjun details three-part framework for personal AI agents
Jiang Guanjun, AI algorithm lead at Alibaba’s Qwen division, said personal AI agents need coordinated development of models, memory and ecosystems. Qwen has mapped user needs across more than 30 leading industries and identified over 8,000 model capability requirements, involving industry experts in evaluation and optimization. He also emphasized understanding users over time and connecting professional tools and services.

What to know
- 01
Jiang said memory should connect past information with current needs rather than simply accumulate conversations, helping limit noise and inference costs.
- 02
Qwen is building a model-centered open service system integrating specialist knowledge, tools, industry skills and third-party agents, while connecting e-commerce, local services, and expertise in healthcare and education.
Why it matters
The framework provides technical context for how Qwen organizes models, memory and service connections around individual users’ needs.
AI-assisted brief. Read the source for full details. Editorial & sourcing policy

