We like to 🚢 fast

Features

1. Privacy friendly: we never touch your database

Our architecture is designed to never touch your database. We only need metadata about your database (the tables and columns it has) to generate queries. These queries are run on your own infrastructure using our easy to use libraries, so the data never ever leaves your servers.

2. Integrates with almost everything

We integrate with all major databases and data warehouses, including Postgres, MySQL, Redshift, Snowflake, BigQuery, MongoDB, Azure Synapse, and SQL Server. We also have integrations for all places where your users might ask questions. Web apps, Mobile apps, Slack bots, Telegram/Whatsapp bots are all supported.

We also have a REST API that you can use to integrate with any tool of your choice.

3. Easily fine-tune the model to your use-case

Our integrations take less than 30 minutes, and you can fine-tune the model to your use-case in an hour. We also have a powerful UI that lets you fine-tune the model without writing any code.

Changelog

2023-02-23 - Self-serve Notion integration 🤝

2023-02-17 - Performance and stability improvements, fuzzy caching 💪

2023-02-10 - Cookbook, platform API, postman collection 📖

2023-02-05 - Azure SQL and Synapse integration 🌞

2023-02-04 - Query explainability and more stable queries 👨‍🏫

2023-02-03 - Citations for unstructured data queries 📖

2023-02-02 - Implemented query analytics 🫘

2023-02-01 - Snowflake integration ❄️

2023-01-31 - Chatbot integrations for Telegram and Whatsapp 🤖

2023-01-30 - Slack integration 🤖

2023-01-29 - Microservice creation scripts and docs 📝

2023-01-28 - Notion integration 📝

2023-01-27 - Redshift integration 🚀

2023-01-26 - Support for unstructured data queries 📖

2023-01-25 - Support for memoryful querying 🧠

2023-01-24 - Experimental auto visualization 📈

2023-01-23 - Front-end components for React 📦

2023-01-22 - Revamped onboarding flow 🎉

2023-01-21 - MySQL integrations 🐬

2023-01-20 - NodeJS library for easier integrations 🐕

2023-01-19 - Self-adaptive learning, Defog automatically corrects its mistakes 🧠

2023-01-18 - Metadata input and editing via Google sheets ✍️

2023-01-17 - BigQuery integration 🔎

2023-01-16 - MongoDB integration 🍃

2023-01-15 - Rudimentary reasoning engine 🧠

2023-01-14 - More secure queries with hard-filters 🔒

2023-01-13 - Python library for easier integrations 🐍

2023-01-12 - PostgreSQL support 🐘

2023-01-11 - Multi-table queries and complex joins 🤝

2023-01-10 - Custom metrics and domain-specific jargon 🗣️

2023-01-09 - Auto-prefilled column metadata descriptions 📝

2023-01-08 - Experimental Google Sheets Integration 📈

2023-01-07 - Multi lingual queries in 20+ languages 🌍