Agentic Coding Unleashed From Autocomplete to Autonomous
Wentao Liu, Zeguo Mao, Xianxiong Luo
Chinese Session 2026-08-09 14:30 GMT+8 (ROOM : JingMing Hall) #agenticcodingAgentic Coding isn’t just about smarter autocomplete—it’s a paradigm shift where AI‑driven tools proactively understand context, make autonomous decisions, and collaborate across platforms to build, integrate, and maintain complex systems. In this talk, we’ll move beyond simple code generation to demonstrate how Cursor (the AI‑native editor), Trae (the AI‑powered testing & deployment agent), and OpenClaw (the embeddable AI assistant) work together as a cohesive, multi‑agent development squad. We’ll walk through a real‑world scenario: autonomously designing, implementing, testing, and deploying an intelligent chat‑room integration, showcasing how Agentic Coding reduces human‑to‑tool friction and elevates developers from coders to system architects.
Speakers:

Wentao Liu: Manager of omfoss.com
Wentao Liu has been a passionate contributor to the Apache Software Foundation for many years. With over two decades of experience in the Linux and open‑source ecosystem, he has played a pivotal role in advancing the OpenMeetings project—from core development and community building to real‑world enterprise adoption.
His contributions include: Delivering talks and workshops about Apache OpenMeetings at major open‑source conferences such as ApacheCon, FOSSASIA, FOSDEM, and CommunityOverCode, helping to evangelize the platform’s capabilities to a global audience. Founding a company that builds commercial video‑conference appliances based on Apache OpenMeetings, demonstrating how the project can be tailored for secure, scalable, on‑premise deployments. Driving the integration of modern authentication (OAuth2, LDAP), virtual backgrounds, mobile applications, and clustering features that make OpenMeetings ready for enterprise use.

Zeguo Mao: Sophomore
Zeguo Mao is a sophomore at Central South University of Forestry and Technology, majoring in computer science and Technology. With a strong foundation in C, C++, Java, and web technologies including SpringBoot, HTML/CSS, and JavaScript, Mao has cultivated a practical, project-driven approach to building interactive systems. His technical curiosity extends deeply into the realm of Artificial Intelligence, where he possesses a solid understanding of machine learning and deep learning algorithms.
Mao’s hands-on experience with intelligent devices, particularly the development of voice-interaction applications for Rokid smart glasses, gives him a unique perspective on human-computer collaboration. He led the “Universal Manual” project, creating a fully voice-controlled object recognition and intelligent explanation system that demonstrates the seamless integration of AR hardware with AI-driven context awareness. Additionally, as the principal investigator for the “Betel Nut Automatic Picking Robot” project, Mao was responsible for market research, feasibility analysis, and the comprehensive technical design of an autonomous system.

Xianxiong Luo: organizer of GitCoffee Open Source Community
Xianxiong is a seasoned technology leader with 17+ years of hands-on experience in software development, enterprise architecture, and technical team management across E-commerce, Ride-hailing, and Open Source ecosystems. Former Product Manager for the Membership Business Line at Better Life Group and former Architect & Technical Director at Guangcai Shenzhou Group. Currently serving as the Head of Yifei Open Source and Lead Maintainer of GitCoffee Open Source Community.
Recognized as a 2024 & 2025 Outstanding Partner of the OpenTiny Open Source Community (under Huawei Cloud). Deeply focused on bridging open-source community building with enterprise-grade AI applications. Core technical expertise spans Java Backend Architecture, Full-Stack Development, and AI Agent Implementation.