Mintive
Mintive builds infrastructure for the agentic enterprise: a governed platform that gives AI agents the models, tools, and company data they need while enforcing role-based access and recording every action. It governs both tool calls and model calls so enterprises can deploy agents safely alongside employees.
At a Glance
- Enterprise security, IT and platform engineering teams
- Companies deploying AI agents against internal systems
- Healthcare and other regulated industries
- Mid-market and enterprise organizations from 1 to 10,000+ users
AI Tools by Mintive
(1)
Mintive
AI Agent Governance Platform
Discussions
No discussions yet
Be the first to start a discussion about Mintive
Latest News
Mintive announces $11M total funding, including a $7M seed led by Cowboy Ventures
MintMCP becomes Mintive and introduces its integrated agentic-enterprise infrastructure platform
Explains how MCP’s enterprise authorization extension changes AI-agent access control
Okta partners with Mintive to govern how agents connect to enterprise apps
Products & Services
Central catalog and gateway for approved MCP servers; bundles servers by role, manages OAuth/API credentials, applies tool-level controls, supports hosted and custom servers, and logs calls.
Routes model calls through one endpoint, applies per-team budgets and approved-model policies, meters usage and cost, and supports OpenAI, Anthropic, Fireworks and other providers.
Agent-scoped identities and bundles with delegated per-connection credentials, least-privilege tool access, short-lived tokens and independent revocation.
Centrally maintained screening policy for MCP tool calls and results that detects prompt injection, secrets, PII and harmful content; supports monitoring and enforcement modes.
Market Position
Mintive positions itself as an agent-governance and identity layer spanning MCP tools, LLM model calls and agent execution, rather than only an MCP proxy. It competes with MCP/agent gateways such as Runlayer and Natoma, model gateways such as LiteLLM and Portkey, and broader platform or identity offerings from Cloudflare, Stytch, WorkOS, Okta, AWS, Microsoft, OpenAI and Anthropic. Its differentiators are quick deployment, per-agent identity and delegated credentials, centralized tool and model policy, inline Mint Guard screening, and unified auditability.
Leadership
Founders
Jiquan Ngiam
Co-founder and CEO; founding engineer and employee No. 1 at Coursera, hired by Andrew Ng, and later worked on neural networks at Google Brain.
Vijay Vasudevan
Co-founder and CTO; previously at Google Brain and one of the co-creators of TensorFlow.
Bozhi See
Co-founder; previously worked at Brilliant Home and Shopkick.
Phillip Bensaid
Co-founder; previously worked at Disney+ and Amazon.
Executive Team
Jiquan Ngiam
Co-founder & CEO
Founding engineer and first employee at Coursera; later worked on neural networks at Google Brain.
Vijay Vasudevan
Co-founder & CTO
Former Google Brain engineer and co-creator of TensorFlow.
Founding Story
Mintive was started after the founders saw early enterprise agents reach production with broad access to company systems and no clear record of which agent took which action. Formerly MintMCP, the company’s initial vision was to treat every agent like an employee: give it its own identity, least-privilege access, delegated credentials, and an auditable record of its work.
Business Model
Revenue Model
Quote-based enterprise SaaS combining per-user licensing based on active AI-agent users with platform fees that scale with usage and team size; enterprise SLAs, dedicated support and flexible deployment are available.
Pricing Tiers
Pricing scales with team size; the quote flow supports teams of 1–100, 101–1,000, 1,001–9,999 and 10,000+.
Target Markets
- Enterprise security, IT and platform engineering teams
- Companies deploying AI agents against internal systems
- Healthcare and other regulated industries
- Mid-market and enterprise organizations from 1 to 10,000+ users
- Enterprise governance of MCP servers and AI-tool access
- Deploying autonomous agents with separate identities and scoped permissions
- Monitoring and auditing AI-agent activity for security, compliance and incident response
- Controlling model usage and spend across teams
- Running AI coworkers in Slack and Teams for scheduled or proactive work
- Protecting healthcare, financial and other sensitive enterprise workflows from data leakage and prompt injection
- Ro
- Modern Treasury
- University of Michigan
- Coursera