AICon Beijing Station: Discussing the Implementation of Database Copilot in the Database Field

This topic has been translated from a Chinese forum by GPT and might contain errors.

Original topic: AICon 北京站,聊聊 Database Copilot 在数据库领域的落地

| username: YY-ha

AICon Global Artificial Intelligence Development and Application Conference and Large Model Application Ecosystem Exhibition is here!

The conference will be held from May 17 to 18 at the Beijing Marriott Hotel Changping. This year’s conference brings together major internet companies such as Alibaba, Huawei, ByteDance, Xiaomi, NetEase, Google, Tencent, as well as leading AI companies like Dark Side of the Moon, iFlytek, and Zhipu, and numerous application vendors like OPPO, vivo, and Himalaya to discuss practical experiences in various scenarios such as Agent, multimodal, RAG, Copilot application building, intelligent office, and going global.

PingCAP Topic Preview

Based on the understanding of this conference’s theme, Li Li, head of PingCAP AI Lab, will share “Database Copilot Implementation in the Database Field” in the Copilot application building practice session.

Database systems, as a complex technical product, have always been a major challenge in management, operation, and diagnosis. In the past, we relied on expert systems, time-series algorithms, reinforcement learning, and other technical means for database optimization, but these methods often only work for specific problems and are difficult to comprehensively cover all aspects of the database system. In this presentation, we will explore the key issues faced in the database field and introduce the technical capabilities required to solve these problems from the perspective of Copilot. We will explain in detail how to achieve practical solutions to problems through LLM’s Flow and Agent applications, and how to use business data in the production environment for collection and feedback to continuously improve Copilot’s performance. Finally, we will demonstrate how Database Copilot solves specific business problems.

The business database and vector database used by Copilot are both TiDB Serverless, which quickly builds various application scenarios with a simplified tech stack. Interested friends can click the following card to check out Hao Huo’s recent sharing at QCon Beijing on the development and application of RAG AI Chatbot based on TiDB Vector.

Full agenda of this AICon Beijing session:

Event Registration

Conference website: AICon_全球人工智能开发与应用大会_InfoQ技术大会

April QCon Review & TiDB Vector Experience Application