Vectara and Quepid

Connecting Vectara to Quepid

We have integrated a preview version of integrating results from Vectara into Quepid.

Prerequisites

You will need a Vectara account and have your Customer ID ready. We also assume that you have an API key (OAuth2 integration is not yet supported) and know the corpus ID of the corpus you want to query.

Using Vectara

When creating a new case using the Quepid wizard dialogs pick the Vectara search engine and click on the "Advanced" option button. There you need to configure "Custom Headers" using your customer ID and API key as in the screenshot below. Vectara requires the headers customer-id and x-api-key to be set to the respective values.

Then finish the case creation wizard and wait for the case to load.

In the Case view, click on "Tune Relevance" to change the query that is sent to Vectara in the "Query Sandbox" editor:

• Set the customerId parameter to your customer id.
• Set the corpusId parameter to your corpus id.

By default, Quepid will extract the use the document's id and display the document's title. If you have further metadata added to your Vectara documents these fields are available as well but need to be configured in the Settings section under Displayed Fields.