Stop advertising to companies.
Start advertising to the people who decide.

Kinobi's agents read the public web to find companies in a buying window — then run your campaigns to the exact buying committee and the people they trust, across LinkedIn, Google, Meta, and Reddit. Priced on booked meetings, not clicks.

# find warm paths into any target account
kinobi search "enterprise, hiring CX, SAP or Salesforce"

graph loaded · researched fresh
12 matched · 47 warm paths

[1] Dell 238 tech · Salesforce · 9 paths
[2] Best Buy 118 tech · Salesforce · 7 paths
[3] Pfizer 98 tech · ServiceNow · 8 paths
[4] 3M 88 tech · Oracle · 5 paths
... 8 more

The model

One graph in. Every channel out.

Kinobi folds every public source into one company graph, reads the why-now signals that put a company in a buying window, then builds ad audiences of the actual decision-makers — and those they trust — and runs the campaigns across LinkedIn, Google, Meta, and Reddit.

How Kinobi turns public data into ad spend Six public data sources fan into a single company graph; the graph surfaces why-now buying signals that feed the Kinobi DSP engine, which builds contact-level audiences and runs campaigns across four ad platforms — LinkedIn, Google, Meta, and Reddit — optimizing to booked meetings. Public data sources The Kinobi engine Ad platforms contact-level audiences Web crawlsite · stack · content Job postsroles · hiring pace SEC filingsfinancials · org Packagesnpm · PyPI usage DNS recordsinfra fingerprint App storesmobile footprint Company graph why-now signals DSP engine builds contact-level audiences · optimizes to outcomes · LinkedIn Google Meta Reddit Optimized to booked meetings & closed deals — not clicks
Public sources become one graph. The graph surfaces who's in a buying window and who decides — then the engine builds contact-level audiences and runs the campaigns for you across LinkedIn, Google, Meta, and Reddit, optimizing to booked meetings and closed deals.

Fresh, not stale

We don't sell a database. We build yours.

Every campaign starts with fresh research. Kinobi's agents read the public web about your market — every company, every decision maker, every signal — current as of this week, not scraped into a database last year.

Who decides

Map the buying committee — and the people they trust.

Kinobi maps the decision makers at every target account, plus the board ties, former colleagues, and investors they trust — and turns that circle into the audience your campaigns reach. When there's a warm path in, you get that too.

Your Network Shared Context Target Account · Dell
Sarah Chen → Alex Rivera, VP Cloud
Shared board seat at VMware · 3yr overlap
94/100

Why now

Companies tell you when they're buying — if you're reading.

Hiring surges, tech-stack changes, fresh funding, new filings, infrastructure moves — Kinobi's agents read every signal and flag the accounts in a buying window, so your campaigns land while the decision is still open.

Dell dell.com
change detected
Security
Proofpoint CrowdStrike Palo Alto Symantec
Infrastructure
Akamai AWS Kubernetes Terraform
Sales & Marketing
Salesforce Marketo Drift 6sense
Data & Analytics
Snowflake Databricks Tableau
Palo Alto added · Symantec no longer detected · 6sense added
Pfizer pfizer.com
change detected
Cloud & Infrastructure
Google Cloud Cloudflare Azure Databricks
Security
Okta CyberArk Zscaler SailPoint
Productivity
Microsoft 365 ServiceNow Jira monday.com
Data & AI
Snowflake Veeva Dataiku
monday.com added · Jira no longer detected · Databricks added

Signals API

Hiring signals tell you when to strike.

Track open roles, hiring velocity, and department growth. Your agent uses this to time outreach perfectly.

Dell · 84 open roles ↑ 38% (30d)
RoleDeptLevelLocationPosted
VP Cloud Infrastructure
Eng
VP
Austin
1d
Sr. AI/ML Engineer
Eng
Senior
Remote
3d
Enterprise Account Director
Sales
Director
New York
4d
Platform Security Lead
Eng
Lead
Austin
7d
Product Manager, Edge
Product
Senior
Round Rock
10d
Sr. Data Platform Eng
Eng
Senior
Remote
12d
Solutions Architect
Sales
Senior
Chicago
14d
Best Buy · 62 open roles ↑ 51% (30d)
RoleDeptLevelLocationPosted
Head of Digital Experience
CX
Director
Minneapolis
2d
Sr. E-Commerce Engineer
Eng
Senior
Remote
4d
Supply Chain Data Analyst
Ops
Senior
Minneapolis
6d
Staff Backend Engineer
Eng
Staff
Seattle
8d
Revenue Operations Lead
Ops
Lead
Minneapolis
11d
Product Manager, Mobile
Product
Senior
Remote
13d
Enterprise Account Exec
Sales
Senior
Chicago
15d

Graph API

Query the relationship graph programmatically.

5 connectors, 4 shared contexts, 7 targets — your agent finds the strongest path through the graph automatically.

Your Network Shared Context Target Account · Dell
Sarah Chen → Alex Rivera, VP Cloud
Shared board seat at VMware · 3yr overlap
94/100

MCP Server

Drop-in intelligence for Claude Code, Cursor, and any MCP client.

Start the MCP server and your AI agent gets access to every Kinobi tool — search, enrich, signals, graph, and intro — with zero integration work.

# start the MCP server
$ kinobi mcp start
MCP server running on stdio
tools: search enrich signals graph intro
 
# Claude Code auto-discovers tools
> Research Dell and find the best intro path
 
Using tool: kinobi.search
9 warm paths found
Using tool: kinobi.enrich
Akamai, Salesforce, Snowflake
Using tool: kinobi.signals
84 open roles · Eng +38%
 
Drafting intro request to Sarah Chen...

Platform

Everything your AI agent needs.

Six public sources, one graph, five APIs. The same graph that powers Kinobi's campaign engine is yours to query — board affiliations, investor networks, alumni ties, every path scored.

Graph API

Query the relationship graph. Every path scored 0–100 on shared context and recency.

Enrich API

Full tech stack by category with change detection. Know when a competitor's contract expires.

Signals API

Open roles, hiring velocity, department growth. Time your outreach to hiring surges.

MCP Server

Drop-in tool server for Claude Code, Cursor, and any MCP-compatible client.

Structured JSON

Every endpoint returns typed, machine-readable output. Pipe it anywhere.

from public data to booked meetings — made for machines ... and humans ;)