On September 9, the 94th Greater Bay Area FinTech Salon — Intelligent Operations Special Session — was held at the Shenzhen International Fintech Town. The Greater Bay Area FinTech Salon is a flagship event under the Shenzhen–Hong Kong–Macao financial technology exchange mechanism. This session, hosted by the Shenzhen Fintech Association and co-organized by Bonree, brought together technology leaders from securities, insurance, and banking institutions across Shenzhen, Hong Kong, and Macao to discuss cloud-native architecture, automated operations, and the evolution of SRE capabilities — offering a multifaceted view of how intelligent operations are taking shape across the financial industry.
Drawing on real-world deployments of its integrated observability platform, Bonree ONE, Bonree presented a session titled "AI-Native Observability: Exploring New Paths for Intelligent Transformation." The presentation offered a systematic look at the challenges and solutions facing financial-sector IT operations in the era of large language models, introducing a complete methodology built on a unified observability data model, powered by AI operations agents, and safeguarded by a trustworthy and controllable governance framework — providing a reusable practice framework for financial institutions across the Greater Bay Area pursuing intelligent operations.
Helping AI Understand the Enterprise: The Value of Multi-Source Data Fusion in Bonree ONE
Large language models now possess the ability to understand, reason, and act — AI is evolving from a tool into an intelligent partner. Yet most enterprises' operations teams still rely on manually monitoring metrics, analyzing logs by hand, depending on expert experience, and responding only after incidents occur — resulting in slow response times, low efficiency, and high costs. The root cause lies in the growing complexity of enterprise IT systems: data is fragmented, standards are inconsistent, and relationships between entities are broken, making it difficult for AI to truly "understand" how an enterprise is running.
Bonree ONE is built around entity relationships, fusing logs, metrics, traces, and events from multiple sources into a single cognitive map of enterprise operations spanning the user experience layer, application service layer, resource layer, and infrastructure layer. This helps customers break down data silos and establish end-to-end, full-stack observability. Only when AI truly understands the real connections between business and IT can intelligent operations be built on a reliable foundation.
Helping AI Do the Work Well: The Value of Agentic Execution in Bonree ONE
Bonree ONE's Sage AI uses a coordinated model of a primary agent and sub-agents: the primary agent is responsible for understanding tasks, autonomous planning, invoking capabilities, executing actions, and providing feedback, while sub-agents focus on specialized tasks such as log analysis, change analysis, and root cause analysis. The platform covers typical scenarios including development and testing, release changes, routine inspection and maintenance, emergency recovery, disaster-recovery drills, and operational governance — helping operations teams move from isolated point tools to end-to-end task closure. Bonree ONE also features a built-in AI-driven operations engine that combines adaptive deterministic root cause analysis with intelligent alert noise reduction, helping teams quickly pinpoint root causes amid massive volumes of alerts and reducing MTTR from hours to minutes.
Helping AI Be Trusted in Production: The Governance and Trustworthiness Value of Bonree ONE
In industries with high requirements for security, compliance, and control, whether AI can be deployed in production ultimately comes down to governance and trust. Bonree summarizes its AI governance framework around four dimensions: controllable, auditable, traceable, and governable. Every action is executed within defined permission boundaries, with full operational trails and traceable paths; task sequences and decision processes can be reconstructed; and AI capabilities, skills, and tools are all incorporated into a unified enterprise governance system. This mechanism ensures that every AI action has clear boundaries, benchmarks, records, and outcomes — easing organizational concerns about losing control of AI and giving enterprises the confidence to bring AI into production.
Turning Capability into Organizational Assets: The Value of Scalable Reuse in Bonree ONE
Bonree ONE's Sage AI allows the response workflows of frontline experts to be codified into reusable agents and skill templates, elevating operational capability from something dependent on individual experience into a standardized, organization-wide asset. The platform provides full lifecycle management — from creation and testing to review and execution — ensuring content safety and command compliance, while maintaining a complete record of operations and an execution evidence chain to make capabilities traceable and reusable. These capabilities have already been put into practice in scenarios such as fault diagnosis, change verification, high-risk action interception, and routine operations automation, meaningfully improving response efficiency and operational stability.
About the Greater Bay Area FinTech Salon
Launched in 2019 by the Shenzhen Fintech Association, the Greater Bay Area FinTech Salon is a premier exchange platform in the fintech space, having welcomed more than 100,000 participants to date. Past guest speakers have come from institutions including the People's Bank of China, China Merchants Bank, Tencent, Huawei, and WeBank. This session was jointly hosted by the Shenzhen Fintech Association and the Shenzhen Capital Market Fintech Committee, with Bonree serving as co-organizer.