JumpCloud is a cloud platform for managing identity, access and devices in one place. It is built for IT teams that want to run users, applications, networks and laptops without keeping directory servers on premises, and it now extends the same controls to non-human identities and AI agents.
The problem it addresses is sprawl: identity held in one place, devices managed in another and privileged access in a third, with gaps between them. JumpCloud keeps all of it in a single cloud service, which matters more as organisations add AI agents that need identities of their own.
At A Glance
| Vendor | JumpCloud |
|---|---|
| Category | Cloud directory, identity, access and device management |
| Identities covered | Employees, administrators, privileged users, devices, non-human identities and AI agents |
| Deployment | Cloud service with a lightweight agent on managed devices |
| Agentic AI | Agentic IAM, AI Gateway, Shadow AI Governance, AI Assistant |
| Customers | Grab, Canva, CoreWeave, MiQ Digital, AkitaBox |
| Certifications | SOC 2 Type 2 examination and ISO 27001 certification |
| Market classification | Directory & Identity Infrastructure; also Workforce IAM, Privileged Access Management |
What JumpCloud Does
Most organisations still tie their identity estate to an on-premises directory, then bolt separate tools on top for sign-in, device management and privileged access. Each tool keeps its own copy of who people are, so joiners, movers and leavers have to be handled several times and mistakes slip through. JumpCloud replaces that stack with one cloud directory that every other control reads from, managed from one admin console. A person, device or agent is defined once, and access to applications, networks, servers and AI tools follows from that single record. The same approach now covers AI agents, each tied to a human owner and managed across its lifecycle alongside people and devices.
Products
Cloud Directory, Cloud LDAP And Cloud RADIUS
The directory is the core of the platform: user and identity management without on-premises servers. It holds users and groups centrally, serves applications and systems that authenticate over LDAP, and handles network and Wi-Fi logons through Cloud RADIUS. Directory Insights keeps a log of activity, including interactions with the AI Assistant.
Single Sign-On, MFA And Conditional Access
This layer controls how people sign in to applications. SSO and MFA cover SAML 2.0 and OpenID Connect applications through a catalogue of pre-configured connectors, and conditional access policies decide when a login is allowed. Users on managed devices can sign in without a password, using identity-bound, hardware-protected access control.
Device Management
Device management gives IT one console for Windows, Apple, Linux and Android endpoints, covering Windows management, Apple MDM, Linux management, Android EMM, patch management and remote access. The agent creates and manages local user accounts and runs commands on Windows, Mac and Linux, and Linux servers can require MFA over SSH.
Privileged Access Management
PAM, expanded through the VaultOne acquisition, is for teams that need to control administrator sessions to infrastructure. A browser-based gateway brokers access to cloud providers, databases, servers and applications without a VPN, and privileged sessions are monitored and recorded in real time.
Password Vault And SaaS Management
The password vault stores shared and personal credentials. SaaS management helps IT see which cloud applications staff are using, control who can reach them and cut unused licences, through SaaS Discovery, SaaS Access Control and SaaS Spend Optimization.
Agentic IAM
Agentic IAM brings AI agents under the same identity controls as staff. Its AI Gateway is the central place to register human users, non-human identities and agents. An agent registry lets administrators authenticate, group, manage and audit agents separately from users, and each agent has a human owner, so it can be disabled without touching the owner’s account. PAM rules extend to agent workflows, with risk monitoring, session isolation, device health checks and audit logs.
JumpCloud AI Assistant
The AI Assistant lets administrators run the platform through chat. It answers questions about users, devices, apps, assets and certificate status, carries out bulk actions, and writes command scripts and dynamic group logic. Actions wait for human approval before they run.
Shadow AI Governance
Shadow AI Governance is for IT and security teams that need to know which AI tools staff are adopting. It detects GenAI tools, agentic browsers and coding assistants across browsers and managed devices, and offers direct AI connectors, with Cursor as one example. A dashboard shows AI adoption, token spend and departmental activity, policies can warn, block or suggest an approved alternative, and reports support the EU AI Act and SOC 2.
How It Works
JumpCloud runs as a cloud service spread across availability zones and multiple geographic regions. The pieces fit together as follows.
| Component | What it does |
|---|---|
| Cloud directory | Holds users, groups and policies, and is the source for SSO, LDAP, RADIUS and device access. |
| JumpCloud agent | Installed on Windows, Mac and Linux devices. It keeps a long HTTPS poll open to JumpCloud’s servers, so changes arrive in near real time, and runs requests when work is waiting. |
| Protocol services | SAML 2.0 and OIDC for applications, Cloud LDAP for directory-aware systems and Cloud RADIUS for networks. |
| PAM gateway | Brokers and records privileged sessions in the browser. |
| AI Gateway | Supports agent-to-agent and API flows and receives MCP connections that discovery brings under governance. |
Identities And Systems Covered
| Human users | Employees and contractors held in the cloud directory. |
|---|---|
| Privileged users | Administrators reaching cloud providers, databases, servers and applications. |
| Devices | Windows, Apple, Linux and Android endpoints. |
| Non-human identities | All non-human forms of identity, discovered, onboarded and registered as corporate identities in Agentic IAM. |
| AI agents and MCP servers | Agents with their own identity and a human owner, and locally running MCP servers added to a governed inventory. |
| AI tools | GenAI applications, agentic browsers and coding assistants. |
| Applications and networks | SAML and OIDC applications, LDAP-aware systems and RADIUS networks. |
Agentic AI Capabilities
Beyond the agent registry in Agentic IAM, these features apply specifically to AI use:
| Secure MCP connections | Unmanaged MCP set-ups found on staff devices can be routed through the AI Gateway, so connections that began as shadow AI come under the same policy and audit as approved ones. |
|---|---|
| Approved SaaS access | Approved-application rules can be enforced across AI tools as well as for people. |
| Google Cloud and Gemini Enterprise | Agentic IAM is also available built and hosted on Google Cloud, tuned for Gemini Enterprise, with AI Device Trust for Gemini Enterprise sessions. |
Company Background
| Launched | 2013, by Rajat Bhargava, Greg Keller and Larry Middle. |
|---|---|
| Headquarters | Louisville, Colorado, United States. |
| Leadership | Rajat Bhargava (CEO), Greg Keller (co-founder and CTO), Micha Hershman (CMO), Amy Moynihan (Chief People Officer), Shianne Sampson (CRO) and Roland Palmer (CISO). |
| Scale | More than 250,000 reported organisations in over 160 countries. |
| Funding | A reported $400 million or more raised. |
| Acquisitions | Stack Identity, an identity security and access visibility platform (January 2025); VaultOne, a PAM vendor (May 2025); Breez, an identity threat detection, investigation and response platform (October 2025). |
Note: This overview is compiled from JumpCloud’s publicly available information as of 4 November 2026. It is provided for information only. NHI Mgmt Group does not endorse any vendor, and nothing on this page represents the opinions of NHI Mgmt Group.