# Things that make them smart

Learns everything. Forgets nothing.

Upload documents, connect Notion or Google Docs, and choose the source content that becomes shared company knowledge. A 6-layer memory system ensures they never forget a conversation, a decision, or a preference. When one employee learns something new, the whole team benefits through shared knowledge.

An AI employee is only as good as what it knows about your business. Upload documents, connect knowledge bases, and feed internal processes directly to your employees.

Knowledge lives in a 6-layer memory system that prevents forgetting. Short-term context, long-term facts, past experiences, learned procedures, entity relationships, and shared team memory all work together.

When one employee learns something new, the entire team benefits through shared knowledge. Train once, apply everywhere.

Getting knowledge in is meant to feel like handing over a folder. Upload files straight into Training, point at a public web page, or connect the tools where your documentation already lives, including Google Drive, OneDrive, Dropbox, SharePoint, Notion, Confluence, Slack, and support desks like Zendesk and Intercom. More than thirty five connectors ship today, and none of them ask you to export a file and upload it again somewhere else. Small documents are usable within minutes, and a large collection processes in the background while showing you its progress.

You decide what goes in, not the connector. When a source can list its pages, folders, channels, projects, or records, you pick exactly which ones enter training. When a source cannot offer a useful list, Sistava stops and asks you to confirm before it imports everything it can reach. That check matters when the account you connected is a shared drive or an inbox holding far more than you meant to teach anyone.

The knowledge base is not a black box. The Training page lists every source and document alongside what was extracted from it, so you can audit what your employees actually learned rather than trusting that it worked. Remove a document or a whole source and its knowledge goes with it. When a page changes in the original tool, retrain that source or schedule an automatic refresh, so staying current never means starting over.

Trained documents are only half the picture. Memory is the part that builds on its own: each employee keeps a notebook of your preferences and corrections that loads at the start of every conversation, carries decisions forward from earlier threads, and quietly maps the people, clients, and projects you keep mentioning. You can open that memory, read what is stored, correct anything wrong, and delete anything stale. Nothing you upload or connect is pooled with other customers or used to train shared models.

Skills are the third kind of knowing, and they are the reason more knowledge does not mean a slower employee. Every skill sits in the employee's mind as a name and a single line at all times, and the full instructions only load when a task genuinely matches. Give an employee five skills or a hundred and fifty and the working context stays just as clear. If you would rather learn the system than teach it, Sistava Academy is a set of free written courses with a quiz and a downloadable certificate at the end.

## At a Glance

- **6** Memory layers
- **100%** Retention across sessions
- **0** Duplicate training needed

## Benefits

### Deep business knowledge

Upload PDFs, Word docs, and internal wikis, or connect Notion and Google Docs directly. Choose the pages, folders, channels, or records to import before your AI employee builds a semantic understanding of that content.

### Never forgets

Short-term, long-term, episodic, procedural, knowledge graph, and shared team memory work together so your AI employee remembers conversations from last month as clearly as ones from five minutes ago.

### Team-wide learning

Brief your marketing lead on new brand guidelines once and the content writer, designer, and social media manager all absorb it instantly. No duplicate training, no inconsistent information across your workforce.

## How It Works

**Your AI employees learn from your business documents and retain knowledge across a six-layer memory system that keeps them sharp over time.**

Upload your SOPs, product docs, contracts, pricing sheets, and internal guides. Your AI agents read and index everything, then draw on that knowledge in every conversation and task. You control what each employee can see, so a support agent knows your help docs and a sales agent knows your pitch deck, without the two mixing.

The six-layer memory system means your AI employees remember more than just what you uploaded. They retain conversation history, build contextual notes over time, track facts about clients and projects, and pull from shared team knowledge. The longer they work, the more useful they become.

One training run serves the team. What you teach becomes shared knowledge your employees draw on, so you teach once instead of repeating the same upload for every employee, and a customer gets the same answer whichever employee they happen to reach. Source selection is the control: you choose what enters that shared base in the first place. The same knowledge serves chat, email, Slack, voice, and scheduled work, so nobody gets a vaguer answer just because the question arrived through a different door.

## Use Cases

### Train your employee on internal docs

Upload SOPs, product manuals, and policies. Your AI employee reads and retains them, so answers reflect your actual processes rather than generic knowledge.

### Let employees remember past conversations

The 6-layer memory system captures preferences, decisions, and context over time. Employees recall what was discussed last week and build on it, not start fresh every session.

### Share knowledge across the team

Add documents to shared knowledge so every AI employee on the team draws from the same source of truth. One update propagates everywhere, instantly.

### Keep knowledge fresh as your business changes

Update training documents any time. Your AI employee picks up the changes on the next interaction, so they stay current without manual re-configuration. Retrain a single source when its pages change, or schedule an automatic refresh after a completed run so it keeps itself up to date.

## FAQ

### Can I train an AI employee on my own company documents?

Yes. Upload PDFs, Word documents, spreadsheets, slide decks, Markdown, CSV, JSON, or plain text straight into Training and your employees can search them within minutes. You can also point at a public web page or connect a tool where your documentation already lives. No cleanup or special formatting is needed first, because the pipeline extracts the text for you, tables included.

### Which apps can I connect as a knowledge source?

More than thirty five connectors ship today, including Google Drive, OneDrive, Dropbox, SharePoint, Notion, Confluence, Slack, Gmail, Outlook, Jira, Linear, Asana, Zendesk, Intercom, HubSpot, Salesforce, GitHub, and WordPress. Any public URL works as a source too, and file uploads need no connector at all. You connect the account once and training happens in place, with no exporting and re-uploading.

### Will my AI employee actually remember what we talked about last month?

Yes. Alongside the documents you train on, every employee keeps memory that carries across conversations: a notebook of your preferences and corrections that loads at the start of each chat, the decisions and context from earlier threads, and the relationships between the people, clients, and projects you keep mentioning. You can open that memory, read what is stored, fix anything wrong, and remove anything stale.

### What is the difference between training an AI employee and its memory?

Training is deliberate. You hand over documents and connected sources and the employee studies them. Memory is what it picks up from working with you, such as a correction, a preference, or what got decided in a meeting. Training gives an employee your facts and memory gives it your context, and both are searched together, which is why a trained answer still sounds like it came from someone who knows how you work.

### Do I have to train each AI employee separately?

No. What you train becomes shared knowledge that your employees draw on, so you teach once rather than repeating the same upload for every employee. That is also why answers stay consistent when a customer reaches a different employee than last time. You keep control of what enters that shared base by choosing which sources, and which pages, folders, or channels inside them, get imported.

### Is my company data used to train AI models for anyone else?

No. Everything you upload or connect stays inside your workspace and is used only to serve your own team. It is never pooled with other customers and never used to train shared models. Remove a document or an entire source and its knowledge is removed along with it.

### What happens when a document changes or goes out of date?

Retrain that source and the new version replaces the old understanding, or schedule an automatic refresh after a completed training run so it stays current on its own. There is no need to re-upload files by hand or rebuild the knowledge base from scratch. The Training page shows every source and document with what was extracted, so you can always see what an employee is working from.

### Does adding more documents or skills slow the AI employee down?

No. Documents are indexed at training time and the employee retrieves only the pieces relevant to the task in front of it, so a large knowledge base stays fast. Skills behave the same way: each one is always present as a name and one line, and the full instructions load only when a task matches. That is why five skills and a hundred and fifty skills feel identical in conversation.

## Knowledge That Compounds

Most tools forget everything the moment a session ends. Your AI employees carry context forward. They remember what a client said last month, what your refund policy says, and what your team decided in the last planning cycle. That continuity is what makes them useful beyond simple one-off tasks.

Shared team knowledge means your entire AI workforce stays aligned. When one agent learns something important, that knowledge is available across the team. You stop repeating yourself and stop getting inconsistent answers from different agents handling the same domain.

## How It Works

1. **Upload Your Business Docs**: Drop in documents, guides, and reference material. Your AI agents index them immediately and use them in every task.
2. **Memory Builds Over Time**: Conversations, notes, and decisions accumulate across six memory layers, giving each agent lasting context about your business and clients.
3. **Share Knowledge Across Teams**: Critical knowledge flows across your AI workforce so every agent operates with the same ground truth.

> I uploaded our entire product wiki and 200 past support tickets. Within a week the agent was resolving issues I didn't even know were common.
>
> Priya N., Customer Success Lead · SaaS company

## Features in Knowledge

- [Remember Every Conversation](/en/features/knowledge/conversational_memory): Your AI employees remember what happened in past conversations, across your whole workspace, not just the one they are in right now. Preferences, decisions, and facts shared in chat get carried forward automatically, so you never have to repeat yourself to the same employee, or explain the backstory to a different one. It works out of the box, with nothing to set up or turn on.
- [Keep Notes Between Tasks](/en/features/knowledge/persistent_notes): Every employee keeps a personal notebook of what it has learned about you: your preferences, corrections, contacts, and how you like work done. That notebook loads automatically at the start of every conversation, so you never have to repeat yourself. Update a preference once and it sticks for good, across every future task and every channel.
- [Custom Knowledge Training](/en/features/knowledge/training): Feed your company's existing knowledge into your AI employees so they answer questions and make decisions using your real data, not generic guesses. Train from a connected app (Notion, Slack, HubSpot, and 65+ more), a website crawl, an uploaded file, or pasted text. Training is company-wide: every employee shares the same knowledge base and searches it automatically on every conversation.
- [Train from File Uploads](/en/features/knowledge/training_from_file): Upload a PDF, Word doc, Excel sheet, PowerPoint deck, plain text, Markdown, CSV, JSON, or XML file straight into Training, and every employee on your team can search it within minutes. Up to 20 files per run, each up to 72 MB, feed the same shared knowledge base your employees check on every message. No connected app and no crawlable URL required.
- [Academy Courses](/en/features/knowledge/academy_courses): Sistava Academy is a set of free, self-paced courses that teach you how to get real work out of an AI employee. Each course is a short series of written lessons you read in the browser, no video required, followed by a quiz. Pass the quiz and you get a certificate of completion with your name on it, ready to download as a PDF.
- [Shared Team Knowledge](/en/features/knowledge/shared_team_knowledge): What one AI employee learns in conversation, every employee in your workspace can draw on, automatically. Tell your sales lead about a new pricing rule and your support employee already has it the next time a customer asks, with no retraining and no copy-pasting the same context into each employee. Train an employee on a set of documents (a policy, a playbook, a product spec) and that knowledge is searched before every response across the workspace, so new hires start from what the team already knows instead of a blank slate.
- [Train From Any Source](/en/features/knowledge/knowledge_sources): Train your AI employees on the documents and tools your company already uses: upload files, point at any URL, or connect an app like Notion, Google Drive, Slack, or Zendesk. Training happens in place, with no exporting and re-uploading, and it teaches the whole workspace at once, not just one employee.

## Explore

- [Every feature](/en/features)
- [Browse AI employees](/en/agents)
- [Pricing](/en/pricing)