Mintive Raises $7 Million to Simplify Enterprise AI Agent Management

The Founders Behind the Platform

Jiquan Ngiam and Vijay Vasudevan have spent years building the infrastructure that now underpins much of modern AI. Ngiam joined Coursera as its first engineer, working directly under Andrew Ng, before moving to Google Brain to focus on neural networks. Vasudevan co-created TensorFlow during his tenure at Google Brain, the training platform that became the industry standard before PyTorch emerged.

Now the pair have turned their attention to a problem enterprises are only beginning to grapple with: how to manage the expanding population of AI agents that will soon operate across corporate systems. Their answer is Mintive, a platform designed to provision, govern, and monitor autonomous agents at scale.

The company announced it has raised $11 million in total funding, including a $7 million seed round led by Cowboy Ventures. Coatue, Operator Collective, and Hustle Fund participated, alongside individual backers including Okta co-founder Frederic Kerrest, Discovery Loop co-founder Jeff Dean, and OpenAI co-founder Andrej Karpathy.

What Mintive Actually Does

At its core, Mintive provides identity and access management for AI agents, a category that barely existed two years ago. The platform includes an MCP gateway that lets IT teams define which cloud services an agent can reach and what actions it can perform. A separate LLM gateway routes requests to different models and enforces cost caps. The system also handles agent identities that may diverge from the permissions of the employees they serve, plus a monitoring layer that tracks all agent activity across an organisation.

Beyond governance, Mintive allows enterprises to build agents capable of executing long-running tasks autonomously, a step beyond the chatbot interfaces most companies have deployed so far.

The pitch centres on ease of deployment. According to Ngiam, setup requires only a few clicks. "What our customers look at and they are a bit shocked at, is how quick it is and easy to deploy," he said. The platform abstracts away the complexity of sandboxes, harnesses, and memory management, handling those layers in the background.

A Crowded Field with Heavyweight Competition

Mintive is entering a market that has become densely populated in a matter of months. Amazon, Microsoft, OpenAI, and Anthropic all offer agent management capabilities as part of their broader platforms. Point solutions have proliferated as well, from LangChain and New Relic to Rippling. Y Combinator alone has backed more than 40 startups working on some variation of agent infrastructure.

The question for any new entrant is not whether the category matters, it clearly does, but whether there is room for another platform when the hyperscalers and foundation model providers are already bundling similar functionality.

Mintive's argument rests on its founders' experience designing systems for enterprise use. The platform launched in February and has signed over 50 customers, including Ro, Modern Treasury, the University of Michigan, Coursera, Apax, and Stability AI. That traction in eight months suggests the ease-of-use claim resonates, particularly with engineering teams that lack the bandwidth to build bespoke agent infrastructure in-house.

The Discipline of a Smaller Round

Cowboy Ventures managing partner Aileen Lee said the founders' backgrounds were a significant factor in her decision to lead the round, but so was Ngiam's approach to capital. "AI is going to be able to power incredible things, but we need more infrastructure that's going to give us governance and visibility and control," she said.

Lee also highlighted Ngiam's restraint in raising only $11 million at a time when other agent-focused startups are securing nine-figure rounds at unicorn valuations. Prime Intellect, for example, raised $130 million for its Series A at a $1 billion valuation. "He is a grown-up and so he was not a person chasing a 'vanity' round," Lee said. "I think a lot of companies are being set up for failure by chasing round size, rather than, 'how much do I need to actually really move the needle for the company?'"

That philosophy reflects a broader tension in the current AI funding cycle. While capital remains abundant for infrastructure plays, the gap between early revenue and billion-dollar valuations has widened. Companies that raise large sums early often face pressure to justify those valuations before they have established product-market fit, a dynamic that can distort priorities.

The Governance Gap

The proliferation of agent management startups points to a genuine infrastructure gap. As enterprises move from pilots to production deployments, the question of how to govern agent behaviour becomes urgent. Agents that can read from databases, write to CRM systems, and trigger workflows across SaaS platforms require a permission model more granular than traditional identity and access management systems were built to handle.

At Opentechwire, we have tracked a steady drumbeat of enterprise security incidents tied to misconfigured API access and overly permissive service accounts. Agents that operate autonomously amplify that risk. A single misconfigured agent with write access to a financial system or customer database can cause damage at machine speed, far faster than a human operator could.

The challenge for Mintive and its competitors is that this problem is still emerging. Most enterprises are still experimenting with agents in controlled environments. The market for agent governance infrastructure will likely expand significantly over the next 18 months, but it is not yet clear which architectural approach, hyperscaler-bundled, foundation-model-native, or standalone platforms like Mintive, will dominate.

What Comes Next

Mintive's early customer base skews towards technology-forward companies and institutions that have already committed to deploying agents in production. The next test will be whether the platform can appeal to enterprises in regulated industries, where the governance and audit requirements are more stringent.

The involvement of Jeff Dean and Andrej Karpathy as investors signals confidence from figures who have shaped the trajectory of AI infrastructure over the past decade. Their participation may also open doors to technical talent and enterprise buyers who trust their judgement.

But the funding environment for AI infrastructure is becoming more selective. Investors who wrote large cheques in 2023 and early 2024 are now scrutinising revenue growth and unit economics more closely. Mintive's decision to raise a modest seed round may prove advantageous if the market corrects, allowing the company to build without the pressure of a valuation it cannot yet support.

For now, the company is focused on expanding its customer base and refining the platform based on feedback from early adopters. The agent management category is still in its formative stage, and the winners will likely be determined not by who raises the most capital, but by who solves the operational pain points that emerge as enterprises scale their agent deployments.