Grab is redefining how engineers respond to operational incidents by introducing Kinabalu AI Site Reliability Engineering (AI SRE); an AI-assisted platform that brings together fragmented context, automates reasoning, and accelerates the path from alert to action. As we have scaled across transport, deliveries, and financial services, we've encountered new complexities that present us with opportunities to innovate and improve our troubleshooting processes. This article explores why we built Kinabalu AI SRE, the design principles and engineering decisions behind it, and how it’s transforming incident response for teams across Grab.

Grab is a leading technology company in Southeast Asia, offering a wide range of services, including ride-hailing, food delivery, and digital payments. Through its platform, Grab provides convenience, accessibility, and reliability to millions of users across the region. From their blog developers can learn from Grab's innovative approach to technology and business, gaining insights into building scalable, customer-centric platforms that address real-world challenges and improve people's lives.

Grab Tech Blog

Grab built Kinabalu AI SRE, an AI-assisted platform that consolidates fragmented operational context and automates incident diagnostics. The system combines a signal aggregator that ingests data from monitoring tools (Datadog, Kibana, Grafana), a Model Context Protocol toolkit integrating 25 operational tools, and LLM-powered agents that perform auto-triage and root cause analysis. It offers both static diagnostics (comprehensive service health reports from six domain-specific sub-agents) and dynamic chat (real-time queries via Slack/Web UI). The platform aims to reduce time-to-resolution by providing evidence-backed insights, correlating signals across metrics, logs, dependencies, and recent changes, while keeping engineers focused on decision-making rather than context-switching.

Kinabalu AI SRE - Leveraging AI for scalable diagnostics and alert management (Part 1)