Making AI Capability Buildable, Reusable, and Collaborative: Developer Practice with SenseTime Raccoon's Skill and Expert-Team System

Wangqun Pan Senior Product Operations Manager at SenseTime, Head of Developer Ecosystem for the Raccoon family

Long focused on AI product communication, developer ecosystem, and user growth, with an emphasis on the path from capability building to developer awareness, adoption, and co-creation for large-model products. Currently leads brand and operations for large-model products such as SenseTime Office Raccoon, planning developer-facing content, case studies, events, and community operations around real-world scenarios like AI office work, intelligent task execution, and content generation, helping developers understand, invoke, validate, and integrate product capabilities into their own workflows. Skilled at driving the real-world adoption and sustained growth of large-model capabilities through hands-on task breakdowns, developer case co-creation, community feedback loops, and ecosystem partnerships.

Abstract

A new paradigm for large-model applications: moving from "tool calling" to "capability building," so expertise becomes reusable, composable, iterable AI capability units.

Details

Centered on SenseTime Raccoon's Skill and Expert-Team capabilities, this talk explores a new paradigm for large-model applications: moving from "tool calling" to "capability building." It focuses on how developers can turn domain expertise, complex processes, and scenario experience into reusable, composable, iterable AI capability units, and how the expert-team mechanism enables multi-role collaboration and task orchestration. Drawing on real-world practice, it looks at how developers build intelligent workflows for individuals, teams, and industries on top of Raccoon — pushing AI capability from isolated efficiency gains toward a sustainably scalable productivity infrastructure.