What it does
DataFusion performs scans, joins and arithmetic first, allowing the optimizer to narrow the rows before Jev evaluates questions about meaning.
Apps & data pipelines
JevNQL
A Rust query language that mixes ordinary analytics with row-by-row semantic judgments from Jev.
Bookmark: JevNQL
Keep this in your collection.
Leave a noteWhat would you try with this?
: JevNQL
Only you can see your notes.
Screenshot unavailable. Open the experiment ↗
How you can use it
A developer can use this code to search customer records for specific meaning. First, the app filters data by standard numbers like money spent. Then, it sends the remaining text to an outside AI service called TypeSafe. This service answers plain-English questions about the text, like asking if a customer seems frustrated.
The source includes a testing tool that guesses feelings from simple keywords. A developer can run this tool to test queries without an internet connection. To understand subtle details, the app requires an access key. This key connects the software directly to the real TypeSafe service.