Hebbia ships memory, Word plugin, and new data integrations in October release
The AI research platform's monthly update adds persistent memory to its Max agent, a Microsoft Word plugin, and connectors to Daloopa, Intralinks, and Dynamo.

Hebbia published its October product update on Monday, adding persistent memory and version history to its Max agent, a Microsoft Word plugin, and three new third-party data sources covering fundamentals, deal rooms, and pipeline CRM data.
The memory feature lets Max retain context across a user's prior sessions, a capability analysts have had from consumer tools like ChatGPT for over a year but which has been slower to arrive in enterprise research platforms where data residency and compliance review gate every feature. Version history, shipped alongside memory, lets users roll back agent outputs rather than re-running a workflow from scratch.
The Microsoft Word plugin is the more consequential surface move. Research analysts, PE associates, and credit teams write their memos in Word. A plugin that brings Hebbia's retrieval and extraction into the document itself shortens the loop between "find the number" and "cite the number in the IC memo." It also puts Hebbia in direct surface competition with the Word-native agent features Microsoft has been building into Copilot.

On the data side, three integrations landed. Daloopa, which standardizes fundamentals from filings and transcripts into structured feeds, is now available to Hebbia customers. Intralinks, the virtual data room provider used across M&A and private credit, syncs content directly into Hebbia so deal teams work against a live feed rather than a snapshot export. Dynamo, a deal-pipeline CRM common in alternatives, syncs documents into Hebbia with pipeline records attached.
The October release follows a September update that introduced "Headless Hebbia," an API and MCP (Model Context Protocol) server letting firms embed Hebbia's capabilities into their own internal tools rather than driving everything through Hebbia's UI.
Are your experts using AI to cheat?
Try our free demo to find out today.
Powering institutional-grade transcription for expert networks.
INFLXD provides AI-powered, human-edited transcription with sub-1% error rates for the world's leading expert networks and financial research firms.
Visit inflxd.com →Keep reading.

How Expert Networks Pay Consultants Across Borders: A Field Guide to Global Payout Rails
The payment rails a network uses shape who it can recruit, how fast experts come back, and what compliance paperwork sits behind every one-hour call.

How Buy-Side Firms Handle Expert-Network Call Transcripts After the Engagement Ends: 7 Retention and Access Models
A structural map of how multi-managers, long-onlys, and credit shops retain, redact, or purge expert-call transcripts once the research is done.

proSapient and Oliver Wyman publish survey on agentic AI in sales workflows
A joint survey of 100 sales leaders reports productivity gains alongside a 40% rate of AI-related incidents.

