Integration Tutorial April 13, 2026

Sync Airtable Databases as AI Knowledge

Airtable bases hold structured knowledge that is hard to search by meaning -- product catalogs, content calendars, research databases, CRM trackers. REM Labs indexes every record so your AI can query across tables, views, and linked records semantically.

What Gets Synced

The Airtable integration imports records with their full field structure:

Records -- each row becomes a memory with all field values preserved as typed metadata
Field types -- text, number, date, single select, multi-select, checkbox, URL, and attachment fields are all preserved
Linked records -- Airtable linked record fields become knowledge graph edges connecting related records across tables
Long text fields -- rich text and markdown content in long text fields is fully indexed for semantic search
Views -- you can sync specific views to import only filtered/sorted subsets of a table

Connect via Token

In the REM Labs Console, go to Integrations and click Connect Airtable. Paste your Airtable personal access token and select the bases and tables to sync.

Import via API

curl -X POST https://api.remlabs.ai/v1/memory/sync/airtable/import \
  -H "Authorization: Bearer sk-slop-..." \
  -H "Content-Type: application/json" \
  -d '{
  "airtable_token": "patABC123...",
  "base_id": "appXYZ789",
  "tables": ["Products", "Customer Feedback", "Research Notes"],
  "view": "Grid view",
  "sync_mode": "incremental"
}'

Specifying a view imports only records visible in that view, respecting any filters and sorts you have configured. Omit it to import all records in the table.

Linked Records as Graph

Airtable's linked record fields are one of the most powerful features of this integration. When your "Products" table links to "Customer Feedback" records, REM Labs creates graph edges between them. Searching for a product surfaces related customer feedback through graph traversal, even if the feedback text is not semantically similar to your query.

Search Across Bases

curl -X POST https://api.remlabs.ai/v1/integrations/airtable/search \
  -H "Authorization: Bearer sk-slop-..." \
  -H "Content-Type: application/json" \
  -d '{
  "query": "customer complaints about slow loading times",
  "filters": {"table": "Customer Feedback"},
  "limit": 10
}'

This finds records tagged "Performance" and entries describing "page takes forever to render" -- connected by meaning, not by exact keyword match.

Structured Metadata Filtering

Because Airtable fields are stored as typed metadata, you can combine semantic search with structured filters. Search for "enterprise feature requests" filtered to records where Status is "Approved" and Priority is "High" -- getting the precision of database queries with the recall of semantic search.

API Endpoints Reference

Endpoint	Method	Description
`/v1/memory/sync/airtable/import`	POST	Import records from Airtable tables
`/v1/integrations/airtable/webhook`	POST	Configure real-time record sync
`/v1/integrations/airtable/status`	GET	Check sync status and record counts
`/v1/integrations/airtable/search`	POST	Search within the Airtable namespace

Formula fields: Computed formula field values are synced as read-only metadata. Rollup and lookup fields that aggregate data from linked records are also included. See Airtable integration docs.

Make your Airtable bases AI-searchable

Free tier. Semantic search across every record and linked field.

Get started free →