Speaker:

Owen Xiao

Date:

Mon Sep 21, 2026 @ 04:30pm EDT

Date:

Mon Sep 21, 2026

Time:

04:30pm EDT

Location:

https://cmu.zoom.us/j/92568565952?pwd=N8VKf5xfF5aisG9VoEwdNsCp0dCwel.1Zoom

Title:

Apache Doris: Real-Time Analytics and Search Database for AI Agents

System:

Doris

Video:

YouTube

---

Talk Info:

Apache Doris is a real-time analytical and search database built for sub-second, high-concurrency queries over fresh data and data lakes. It has been widely used for customer-facing analytics, data warehousing, and observability. As AI agents move from prototype to production, we see two new categories of workload for analytical databases: Agentic Analytics, where agents autonomously query and explore data on latency budgets; and Agent Observability, where every agent step needs to be traced, indexed, and queried — often as long unstructured text and deeply nested JSON.

In this talk, I'll walk through how Apache Doris is evolving to meet the challenges: from vectorized, MPP query engine that keeps Data Agents fast, to native support for VARIANT JSON and inverted indexes that make searching and analyzing sprawling agent trace data tractable.

This talk is part of the Whatever Andy Found Interesting While Out At Sea.

Bio:

Owen is an Apache Doris PMC Member focusing on the observability and AI area. He designed and implemented the core features for observability and AI, such as inverted index and VARIANT JSON data type. He has over 15 years of experience in architecture design, development, and team management in the fields of big data, database, and AI.

---