Mention — Onboarding that builds itself from knowledge you already have

Onboarding is broken

A Notion page with a list of links. Some outdated. Some missing context. The order is arbitrary.

Questions in Slack. A senior answers for the fourth time. The answer disappears into chat history.

Meetings with undefined terms. Understanding pieced together from context over weeks.

No way for a manager to know what a new hire actually understands versus what they've simply been exposed to.

#engineering

47

JD

Jordan Davis New Hire 9:15 AM

Hey team! First week here. Can someone walk me through our deployment process?

SK

Sarah Kim 9:22 AM

There's a doc in Confluence somewhere... let me find it

ML

Marcus Lee 9:28 AM

Actually we changed the process last sprint. The Confluence doc is outdated now

JD

Jordan Davis New Hire 9:35 AM

Also, what's a "feature flag"? It keeps coming up but I don't see it defined anywhere

AW

Alex Wong 9:41 AM

Oh, you'll need to understand our CI pipeline first before deployment makes sense. That's in... hmm, not sure where

SK

Sarah Kim 9:45 AM

I explained feature flags to the last new hire in a thread somewhere... let me search

JD

Jordan Davis New Hire 10:02 AM

I'll just ask in standup tomorrow I guess. Sorry for all the questions!

Message #engineering

What if your existing knowledge could onboard people automatically?

Mention connects to the tools your team already uses, extracts what matters, and turns it into a guided learning experience.

Detect concepts

AI identifies important terms, tools, and patterns from your docs, Slack, and meetings.

Extract processes

Step-by-step workflows pulled from documents automatically. No manual writing required.

Dependency ordering

Concepts arranged so learners never encounter unexplained terms. Each idea builds on the last.

Generated articles

Clear explanations written using only terms the reader already knows. Always current.

Knowledge checks

Quizzes that verify understanding, not just exposure. Learners explain in their own words.

Q&A on your knowledge

Learners ask questions, AI answers using your team's actual context — not generic info.

Set up onboarding in one session

Authors are curators, not writers. The AI does the extraction and generation.

1

Create an audience

Name who you're onboarding — "New Engineering Hires", "Sales Team", "Design Onboarding".

2

Connect sources

Notion, Confluence, Slack, Google Drive, Zoom. One-click OAuth. Your knowledge stays where it is.

3

Curate, don't write

Review AI-suggested concepts and processes. Accept, rename, or discard. The AI generates the content.

4

Invite & launch

Learners immediately start working through structured learning paths. No docs to write. No courses to build.

Learning Path

Git Workflow Internal Tool

CI Pipeline Process

Feature Flags Domain Concept

Deployment Process Process

Rollback Procedures Process

Incident Response Workflow

What learners see

Learn in order

Concepts presented so each one builds on what came before. No gaps, no circular references. Every explanation uses only terms you've already learned.

Read, understand, prove it

Articles followed by knowledge checks. Not multiple choice — learners explain in their own words. Understanding is verified, not assumed.

Ask anything

Q&A panel with answers grounded in your team's knowledge, not generic info. Like having a patient senior teammate available 24/7.

Self-paced

No deadline pressure. Progress tracked automatically. Managers see exactly where each person is — what they understand, not just what they've been sent.

Works with the tools you already use

Connect once. Knowledge stays current as sources change.

Notion

Confluence

Google Drive

Slack

GitHub

Zoom

Dropbox

Google Meet

Box

Security & privacy at the core

Your team's knowledge is sensitive. We treat it that way.

Minimal retention

Content is processed and discarded. We store generated analyses tied to concepts, not your raw documents.

Selective listening

Only material relevant to recognized concepts is analyzed, reducing noise and risk from general chat.

Scoped access

Per-integration authorization with separate data stores. No cross-database reads.

Auditable events

Event logs trace when concepts changed and why, ensuring full transparency for your team.

What you can do on day one

1

Create an audience for your team

2

Connect one or two knowledge sources

3

Review AI-suggested concepts and processes

4

Invite team members — they start learning immediately

No documentation written. No content authored. No courses built.

Ready to get started?

Start building your onboarding experience today

Start Now

No credit card required to start.

Frequently asked questions

How is this different from an LMS or wiki?

Mention builds structured learning paths from your existing knowledge, with concepts in the right order and quizzes to verify understanding. No courses to create, no pages to write. The content comes from your team's actual docs, conversations, and meetings.

Do I need to write any content?

No. Authors curate what matters — the AI extracts concepts and generates articles from your existing docs, chats, and meetings. Your job is to say "yes, this concept matters" or "no, skip that one."

How does dependency ordering work?

The AI figures out which concepts depend on which, and sequences them so learners always have the foundation before encountering something new. No one reads a term that hasn't been explained yet.

What if our knowledge is spread across many tools?

That's exactly what Mention is built for. Connect Notion, Confluence, Slack, Google Drive, Zoom, and more. The AI synthesizes knowledge across all sources into a unified learning experience.

How do articles stay current?

Articles are generated on demand from the latest source material. When a Notion page is updated or a new Slack discussion happens, the AI incorporates that information the next time an article is generated.

What models do you use?

We use state-of-the-art language models and validators behind the scenes. The exact providers and prompts aren't the product; the structure and guarantees are.