New in Siesta AI: Memory as your organization’s knowledge base

We’re introducing enterprise AI knowledge base capabilities in Siesta AI with Memory: a central place to store long-term, high-signal knowledge such as internal processes, product documentation, policies, FAQs, and team playbooks, then attach it to specific Agents. The operational problem it solves is familiar: answers live across wikis, drives, and old threads, so people spend time searching, re-asking, and rebuilding the same context. When an agent doesn’t have a trusted source of truth, responses become inconsistent, and adoption stalls.

Enterprise AI knowledge base: why Memory matters

Memory is designed for organizations where knowledge is not “nice to have,” it’s part of daily execution. Instead of relying on tribal memory or yet another document repository no one maintains, you create a structured space agents can reference first. This keeps guidance consistent across teams and makes updates straightforward: change the source once, and every agent connected to it reflects the new reality.

Collections: separate knowledge by team, project, or sensitivity

Knowledge in Memory is organized into Collections, which work as thematic folders. This lets you split information in a way that maps to how the organization operates: Marketing enablement, Sales playbooks, HR policies, Security standards, or a project-specific rollout.

Collections include practical administration controls. You can create a new collection from the dropdown menu, then manage it using built-in actions to set permissions, rename, or delete. Collections can be private or shared within a team or across the organization, which matters in real environments where some knowledge should be broadly accessible while other content needs tighter control.

A useful pattern is to keep shared baseline collections for company-wide workflows, and team-private collections for details that only certain roles should use day to day. This structure helps prevent agents from mixing policies across teams and gives you a clean way to govern what each agent is allowed to pull from your enterprise AI knowledge base.

Pages and nesting: build maintainable playbooks, not document sprawl

Inside each collection, you create Pages. Pages can be nested into subpages, which means you can build a hierarchy like an internal wiki, not just a flat list of notes.

Nesting is what makes Memory maintainable over time. A top-level page like “Customer Support Handbook” can have subpages for tone of voice, escalation paths, refund policy, known issues, and system-specific procedures. When something changes, you update one place. You avoid the common failure mode where the real policy exists in multiple slightly different versions across tools.

A content editor that supports structure and retrieval

Each page opens an editor where you can format content clearly using headings (H1–H6), lists, links, and code blocks. This is not just about aesthetics. Well-structured pages are easier for humans to scan, and easier for agents to search semantically.

A practical recommendation is to write Memory pages the way you’d want to read them during a busy moment: short sections, explicit definitions, decision rules, and links to the system of record. The clearer the structure, the fewer almost-correct answers you get when someone asks a detailed question across your enterprise AI knowledge base.

Link Memory to Agents, and keep knowledge current without rework

Memory becomes operational when you attach a collection or page to an Agent. Once linked, the agent will search those sources first for each query, draw facts from them, and use your company terminology. Updates in Memory reflect immediately for all connected agents, so there’s no need to reconfigure agents every time a process changes.

This is the difference between an agent that answers based on general patterns, and an agent that answers based on how your company actually works today. It also enables a clean single source of truth workflow: the doc is the policy, and the agent is the interface that helps people apply it.

Example: company contacts that stay current

A strong starting point is a Company Contacts page. Most teams already have this information somewhere, but it’s usually fragmented across org charts, spreadsheets, and old messages. In Memory, you can capture contacts in a structured way: who owns a domain, who the backups are, and where to reach them.

Attach that page to an internal help agent, and employees can ask questions like “Who owns vendor onboarding?”, “Who is the backup for payroll approvals?”, or “Where do I escalate a customer issue?”. When a new employee joins or responsibilities change, you update the page once, and every agent using it reflects the updated structure immediately.

Availability and how to get started

Memory is available in Siesta AI now. Create a collection for a team, add a few pages (processes, FAQs, definitions), and attach the relevant collections or pages to the agents that need them.

Get started here: https://docs.siesta.ai/memory