An AI tool turns private data questions into dashboards in the browser

Eatmydata.ai was presented as a tool that turns data questions into SQL and dashboards without sending the data outside the browser. When a user asks a question, AI agents create multiple SQL queries for in-browser SQLite and write dashboard configuration code without seeing the results. The data is sanitized and obfuscated before anything is sent to a remote LLM, and the dashboard runs inside the browser. It was also described as a local-first, browser-only data exploration tool released under the MIT license.

Key points

  • The tool says data analysis runs in the browser with no separate backend.
  • AI creates SQL queries and dashboard configuration, but does not see the query results.
  • It uses SQLite, a local semantic index, and vector search inside the browser.
  • The remote LLM is said to receive only sanitized and obfuscated information.
  • It targets quick ad hoc data analysis and is released under the MIT license.

Quick term guide

Configuration
The settings that tell a program how to run correctly.
local-first
An app design where your data is mainly stored and controlled on your own device.
MIT license
A permissive license that usually allows broad reuse if basic conditions are kept.
Solo developer
An individual who handles all parts of creating a project or product alone.
developers
Developers are people who build software, apps, or websites.
data analysis
The work of looking at data to find useful patterns or answers.
semantic index
A search helper that finds things by meaning, not only by exact words.
vector search
A search method that finds text with similar meaning, not only the same words.

Sources covering this story (2)

Read original