Loading video player...
"Keeping APIs reliable across distributed services is challenging, and many teams only find out something is broken when users complain. This talk shares a practical approach to making observability part of the API lifecycle instead of treating it as an afterthought. It covers building a unified telemetry pipeline using OpenTelemetry, defining meaningful SLIs and SLOs, and applying AI for anomaly detection, predictive incident alerts, and faster root-cause analysis. The focus is on turning scattered telemetry into clear insights that help engineers respond quickly, reduce MTTR, and build APIs that stay reliable at scale. More details: https://confengine.com/conferences/apidays-india-2025/proposal/23578 Conference Link: https://www.apidays.global/archive/2025/india"