# Institutional knowledge doesn't scale: Building an agentic data analyst

DevFeed: [Institutional knowledge doesn't scale: Building an agentic data analyst](<https://devfeed.tech/articles/institutional-knowledge-doesn-t-scale-building-an-agentic-data-analyst-11588.md>)

Original publisher: [Read original article](<https://incident.io/blog/agentic-data-analyst-pt-i>)

Author: Navo Das

Published: 2026-08-03T10:45:52Z

Content type: article

Language: en

Sources: [The incident.io Blog](<https://devfeed.tech/sources/the-incident-io-blog.md>)

Topics: [data](<https://devfeed.tech/topics/data.md>), [semantic-layer](<https://devfeed.tech/topics/semantic-layer.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [text2sql](<https://devfeed.tech/topics/text2sql.md>), [SQL](<https://devfeed.tech/topics/sql.md>), [Databases](<https://devfeed.tech/topics/databases.md>), [metric-standardization](<https://devfeed.tech/topics/metric-standardization.md>)

Tags: [agentic](<https://devfeed.tech/tags/agentic.md>), [analytics](<https://devfeed.tech/tags/analytics.md>), [building](<https://devfeed.tech/tags/building.md>), [data](<https://devfeed.tech/tags/data.md>), [incident](<https://devfeed.tech/tags/incident.md>), [incident-channel](<https://devfeed.tech/tags/incident-channel.md>), [incident-management](<https://devfeed.tech/tags/incident-management.md>), [incident-response](<https://devfeed.tech/tags/incident-response.md>), [llm](<https://devfeed.tech/tags/llm.md>), [outage](<https://devfeed.tech/tags/outage.md>), [post-mortem](<https://devfeed.tech/tags/post-mortem.md>), [semantic-layer](<https://devfeed.tech/tags/semantic-layer.md>), [slack-incident](<https://devfeed.tech/tags/slack-incident.md>), [sql](<https://devfeed.tech/tags/sql.md>), [tool](<https://devfeed.tech/tags/tool.md>)

## AI overview

The article explains why dashboard-based self-service analytics and direct LLM access to a data warehouse leave important gaps. It describes building an agentic data analyst, called the "data brain," to distribute institutional data knowledge and help employees ask questions while addressing issues such as canonical joins, filters, metrics, and judgment required to produce correct SQL.

## Source excerpt

Institutional knowledge was always the bottleneck. Here's how we built an agentic data analyst to distribute it more efficiently -- and what happened when we let the whole company ask it questions.