Your brain’s human context layer

HumanKit captures and stores your team’s meetings, calls, conversations and computer activity so you and your agents can query and act on it.

Block A · Tabbed showcase (PostHog style)

Block B · Proof strip

Built with the teams building company brains

Design partnerDesign partnerDesign partnerDesign partner

Block C · Statement + stat band

Brains read what’s written. HumanKit captures what’s said.

Your brain already connects Slack, email, docs and the CRM. The context that matters most, what people say, isn’t written anywhere.

5 minto your first recorded meeting
130+languages, detected automatically
8+speech engines, the best per language
100%white-label, your notetaker’s name

Block D · Comparison table

Why not a notetaker, or a bot API?

Personal notetakersMeeting bot APIsHumanKit
Every meeting in the company✕ one person’s✓ if you build it✓
Rules set by the company✕✕ build it✓
Speakers named, remembered~ often wrong✕ labels only✓
Calendar sync✓✕ build it✓
Stored and queryable✕ in their app✕ stateless✓ API and MCP
Under your brand✕✓✓

Block E · Split: copy + code

For developers

Live in five minutes

One API key, a connect link and a webhook. Calendar sync, the notetaker and all the processing run on HumanKit.

  • Users connect Google or Microsoft in your product
  • meeting.finished the moment a meeting is ready
  • Transcript, speakers and summary in one read
Connect a calendar
curl -X POST \
  https://api.meetingkit.com/api/v1/calendar_connect_links \
  -H "Authorization: $MEETINGKIT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "workspace_id": "wks_…",
    "member_id": "mem_…",
    "provider": "google",
    "return_url": "https://example.com/connected"
  }'
Who said what
"speakers": [{
  "id": "spk_1",
  "name": "Ana García",
  "email": "ana@acme.com",
  "identified": true
}]

Block F · Bento grid

Every speaker named

Separated even in crosstalk, matched to the people invited, recognised in the next meeting.

130+ languages

Detected automatically, even when a call switches language.

Your brand

Your notetaker’s name, avatar and messages in every call.

Company rules

What gets recorded, per company, person and meeting. Consent built in.

Built for agents

Webhooks the moment anything changes, one read per meeting, and MCP for your agents.

Block G · Two audiences

Company brains

Speech is your richest source of context

Ship your own notetaker instead of depending on someone else’s, with who-said-what you can trust.

Build it in →

AI consultancies

Step one of every AI rollout

Deploy a notetaker across your client’s company today, under your brand, and give your agents what people say.

Set up a customer →

Block H · Roadmap: from meetings to all human context

NowMeetings

Zoom, Meet, Teams

NextCalls

Phone and on-call

NextScreens

Computer activity

LaterIn person

The room, the coffee

Block I · Quote

"I need a notetaker that knows who is who, across the whole company, so my brain can reach anything and anyone."

Founder and CEO, placeholder until a partner agrees to be quoted

Block J · Closing CTA band

Give your brain the human context

Start with meetings today. Free while we build it with our design partners.

Story S1 · Plain essay, bold lead-ins

Every company is building a brain. Agents that read Slack, email, docs and the CRM, and act on them. But most of what a company knows was never written down. It was said: in meetings, on calls, at someone’s desk, over coffee.

That context is lost today. Notetakers capture a slice of it, one person at a time, and often get who said what wrong. Nobody governs them, and the brain never sees the rest.

HumanKit is the human context layer. It captures what people say and do at work, knows who said it, and keeps it, so you and your agents understand what everyone’s doing and can act on it.

It starts with meetings. MeetingKit, our first product, puts your notetaker in every Zoom, Google Meet and Microsoft Teams call, names every speaker in 130+ languages, and hands it all to your brain through one API. Calls, screens and in-person conversations come next.

Story S2 · Numbered chapters

01 · The problem

What people say is the context brains never see

Brains read what’s written. Decisions, objections and promises happen out loud, and personal notetakers keep only fragments, with the wrong names attached.

02 · What we’re building

The human context layer

One place that captures what your team says and does, attributes it to the right person and keeps it, so you and your agents understand what everyone’s doing and can act on it.

03 · Today

MeetingKit

Every meeting in Zoom, Google Meet and Microsoft Teams, under your brand. Speakers named, 130+ languages, the best of 8+ speech engines, one API and MCP. Live in five minutes.

04 · Next

Everywhere work is said

Phone calls, then screens and computer activity, then the room itself: in-person and hybrid conversations.

Story S3 · Narrative with highlights, then a timeline

Your brain already knows what’s in your tools. It doesn’t know what was said in Tuesday’s renewal call, who promised what in the 1:1, or why the deal stalled. That context lives in people, and it disappears the moment a conversation ends.

HumanKit captures it, names everyone in it and keeps it, so you and your agents understand what everyone’s doing and can act on it.

TodayMeetingKit

Every video meeting, speakers named, one API

NextCalls

Phone and on-call conversations

ThenScreens

Computer activity, with consent

LaterIn person

Hybrid rooms and the coffee table

Story S4 · Manifesto, one line at a time

Brains read what’s written.

Work happens out loud.

The meeting, the call, the hallway, the screen.

HumanKit captures it, knows who said it, and keeps it.

Starting with every meeting. Then everywhere else.

Story S5 · Letter from the founders

For years we built transcription and notetakers. Talking to the teams building company brains, we heard the same thing again and again: speech is their most important source of context, and what they get from notetakers isn’t good enough. Above all, who said what.

So we’re building the layer we wished existed: one that captures what people say and do at work, knows who they are, and keeps it, so you and your agents understand what everyone’s doing and can act on it.

MeetingKit is the first piece, and it’s live: every meeting, every speaker, under your brand, through one API. Calls, screens and in-person conversations are next. We’d love to build them with you.

The HumanKit team, placeholder signature