Rohit Patel founds Ndoc and opens an alpha version of its workspace for data analysis and documents
Rohit Patel has founded Ndoc after eight years at Meta. Its AI workspace brings together data analysis, documents and presentations, and provides access to the Python scripts used. A corporate pilot is underway, and the alpha version is opening to a limited number of people from the waiting list.
Rohit Patel, previously a director at Meta Superintelligence Labs, is leaving Meta after eight years and has founded Ndoc. Its AI workspace brings together data analysis, documents and presentations. The service has its first test users, a corporate pilot is underway, and the alpha version is opening to a limited number of customers from the waiting list.
Users can upload a dataset and request an analysis and a write-up of the results. The Stage tab contains presentations, dashboards or small applications; the Narrative tab contains explanations of the results, citations and images; and the Model tab contains the underlying data and Python scripts. The scripts used can be viewed; according to Patel, users can also ask AI to explain a calculation or find the line containing a specific formula.
According to Patel, a document can be exported as a single file with the .ndoc.html extension that opens in any browser. He plans to release the format as open source so that other developers can also build applications for it. This is a plan, not an announcement of a license that has already been released.
Registration is free, but customers are expected to pay according to usage and token consumption. The article does not provide specific rates. Patel hopes to make the beta version available in the coming weeks and plans a full launch next year.
Why it matters
Linking the final presentation to explanations and underlying calculations makes it possible to trace what AI relied on when preparing the document. This is useful when reviewing analyses used to make decisions; access to the scripts alone, however, does not confirm that the results are correct.
Two audiences, two different impacts
What this means
For individuals
When reviewing their own analysis, users can inspect the Python scripts used and ask AI to explain a specific formula. Access to the service remains limited.
For a business
For businesses, the service brings analysis and the preparation of presentation materials together in one workspace. Costs are expected to depend on usage and token consumption, so free registration does not mean free use.
ProcessesCheck the original
Event sources
only one source so far · 1 publisher, 1 independent. We count feeds from the same owner only once.