Prep tasks for AI Context Engine
Get access to the AI Context Engine API
Contact the support team or your Customer Success Director to get the AI Context Engine API access enabled in the data.world organization for which you want to set up AI Context Engine.
Get access to the Developer Chat UI Tool
data.world provides a web tool for developers to test the AI Context Engine during implementation. Contact the data.world team to enable the tool for the accounts involved in building and setting up the AI Context Engine. Use this tool exclusively for testing and building; it is not intended for roll-out to end users.
The tool URL will be like: https://yourcompany.app.data.world/a/demo/yourorg/yourproject/chat/new
Virtualize your data
Ensure your data is virtualized in data.world and is available in a dataset in data.world. These are the database tables/data sources that you query today to answer the questions you identified earlier. In case data virtualization is not set up at this time, you can alternatively start with a csv export that has the information that we need to answer the identified questions.
Create a project and link to dataset with virtualized data
Create a project that will be used for hosting the Index, R2ML, and ontology files.
Connect this project to the dataset that hosts the virtualized data.
Add the ddw-standard-catalog-index.ttl file provided by data.world to the ddw-catalogs dataset.