Data Fabric
What to know
Data Fabric is Appian's abstraction layer over records, data stores, and integrations. A candidate should understand record types as the core modeling unit: their data source (database table, web service, process, or expression), their fields, and their relationships to other record types (one-to-many, many-to-one, many-to-many). Know the distinction between a record (a row) and a record type (the schema plus source). Understand record-type relationships as the mechanism that lets a form show related data without custom queries, and know that record views, filters, and user filters shape what users see.
Traps
A frequent trap is treating a record type's fields as always backed by a single table - a record type can be sourced from a web service or process, so not every field maps to a column. Another is mixing up a relationship (defined between record types) with a foreign key (a database concept). Questions may ask where a value comes from when a relationship is used on a form - the answer is the related record type, resolved at runtime. Watch for confusion between record-level security and field-level visibility.
Rule of thumb
For any data question, ask three things: What is the source? What are the fields? What are the relationships? If a question can't be answered from those three, it's probably testing security or a view.
Worked example
A Customer record type sources from a database table and relates one-to-many to Order. A form displays a customer and its orders by using the relationship customer.orders - no separate query is written; the relationship resolves the join.