Give Your AI CRM Memory with HubSpot + REM

Your CRM holds the full history of every customer relationship -- calls, emails, deal stages, notes. REM Labs syncs HubSpot into AI memory so your agents and bots have persistent context about every contact and deal without you stuffing prompts manually.

What Gets Synced

The HubSpot integration pulls CRM objects and their associated activities:

• Contacts -- name, email, company, lifecycle stage, and all custom properties
• Companies -- company name, domain, industry, size, revenue, and custom properties
• Deals -- deal name, stage, amount, close date, pipeline, and associated contacts/companies
• Notes and activities -- call logs, meeting notes, emails, and tasks associated with each record
• Associations -- contact-to-company, deal-to-contact, and deal-to-company links become knowledge graph edges

Connect via OAuth

In the REM Labs Console, go to Integrations and click Connect HubSpot. Authorize via HubSpot OAuth and select which CRM objects to sync. You can choose contacts only, or include deals, companies, and activities.

Import via API

The pipeline filter lets you target a specific deal pipeline. Incremental mode uses HubSpot's lastmodifieddate property to only re-process records that changed since the last sync.

Query Customer Context

Once synced, your AI has full CRM context. Search by customer name, deal topic, or any aspect of the relationship history.

This surfaces the deal record for Acme Corp, the meeting notes from the pricing call, and the email thread about custom contract terms -- all connected through HubSpot's association graph.

CRM Associations as Knowledge Graph

HubSpot's association model maps directly to REM Labs' entity graph. When a contact is associated with a company and a deal, searching for the contact also surfaces the company context and deal history through graph traversal. Your AI understands the full relationship web, not just isolated records.

API Endpoints Reference