T

he Brief

- Defog transforms your app's insights with ChatGPT for Data, an embeddable and natural language-enabled solution.

- Co-Founder: Medha Basu

- Co-Founder: Rishabh Srivastava

- Investors: Y Combinator (W23)

- History: Just Launched

Launch YC: Defog.ai - ChatGPT for data, embedded in your app

Defog.ai is revolutionizing the way companies provide data insights to their users. Founded by Medha Basu and Rishabh Srivastava, Defog is an intuitive chat-based interface for analytics that can be easily embedded within an app. This innovative solution allows users to ask questions in natural language and instantly receive data and insights.

Image Credits: Defog.ai

‎One of the major problems with traditional insights features within apps is that they take a long time to build and perfect. This is often because SaaS founders and product managers are focused on improving their core product, and insights features are not a top priority. As a result, many companies end up with hastily-built dashboards that provide limited insights to users.

Defog solves this problem by making it easy for companies to offer a chat-based interface for analytics within their app. This allows users to ask questions and receive data and insights in seconds. Some examples of how Defog can be used include fintech companies providing deep insights based on users' transaction history, video meeting tools extracting common complaints from customer calls, and social media platforms replacing generic analytics dashboards with more specific metrics.

Image Credits: Defog.ai

‎To use Defog, companies simply need to provide the metadata of their database schema (if using SQL) or unstructured data corpuses (if using APIs or text corpuses). They can then add Defog's iframe or Javascript plugin to their app and set up visual styling for charts, if desired. Defog is powered by large language models that can search and visualize both structured and unstructured data. Importantly, Defog does not compromise on privacy – the model only requires access to the database schema, not actual customer data.

Early access to Defog is currently available for sign-up. Companies can also try out Defog on their own datasets or on one of the pre-selected ones. If you have any questions or feedback, the Defog team would love to hear from you – you can email them at founders@defog.ai or reach out on Twitter or LinkedIn.

Posted 
March 14, 2025
 in 
Launch
 category
← Back to all posts  

Join Our Newsletter and Get the Latest
Posts to Your Inbox

No spam ever. Read our Privacy Policy
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.