JevMade Sign in
← Back to experiments

SDKs & clients

libsemop

libsemop brings typed AI decisions to C++, including an empty answer when the model is unsure.

Bookmark: libsemop Keep this in your collection.
Leave a noteWhat would you try with this? : libsemop

Only you can see your notes.

Source screenshot of libsemop
SOURCE SCREENSHOTFull screenshot ↗

What it does

This port of Jason Duncan’s Python semantic operators supports filtering, ranking and escalation. Version 0.1.0 is not yet released.

How you can use it

If your app is written in C++, ask your developer to start with the library’s routing example. Give the AI a list of teams to choose from, and keep an empty answer when it is unsure. Your app can then leave that request for a person.

The TypeSafe connection needs an access key for its outside AI service. Several questions can share one input, while filtering checks each item separately. Test uncertainty on examples you know: the confidence setting controls when an answer is withheld, not whether it is correct.

Keep this for later

Sign in to bookmark experiments, guides and videos, and keep notes only you can see.

Continue to sign in

We’ll bring you back to this listing.