Developers hate monotonous work. The entire field of programming is a testament to how hard people work to automate the tedious tasks involved in building things. And, as Vibecoding proves, they’ve even found a way to do most of the work of assembling the product.
So I wasn’t really surprised when I learned that Naïve, which provides the infrastructure that allows AI agents to take on much of the work involved in running a business, signed up over 30,000 developer customers within months of its launch.
Taking Vibecoding a step further, the startup claims that its infrastructure can automate most of the work in starting and running a business (provided you provide an AI agent and the necessary token budget, of course). Package payments, email accounts, phone numbers, cloud infrastructure, storage, and company formation processes in a single API.
Naïve provides prompts that developers can provide to tools such as Cursor, Claude Code, and Codex. These tools can connect to your company’s APIs to provision the infrastructure to establish your business. This allows an agent to coordinate the formation of a US LLC by providing details such as state, industry code, business description, and proposed name, but requires user involvement to complete the KYC/KYB process and make the required payments.
The rest of the configuration can be done by the AI agent, including setting up email inboxes, virtual cards, phone numbers, databases, computing resources, and connections to services like Stripe and QuickBooks. The governance layer promises to help users set budgets, limit agent functionality, and require human approval before sensitive actions are performed. The company also offers templates for businesses such as AI SEO, full-stack SaaS apps, recruiting, accounting, and customer support, as well as a mobile emulator that allows agents to operate smartphone apps on emulated devices.
The appeal of such automation is clearly resonating, as evidenced by the user numbers mentioned above. And Naïve’s annual run-rate revenue has grown 10x in the past six months, reaching low double digits, CEO and co-founder Sean Dorje said.
Building on that traction, the company has now raised $28.5 million in a Series A funding round led by Nexus Venture Partners, TechCrunch has learned exclusively.
Dorje told TechCrunch that his customers are using Naïve to run autonomous businesses such as AI-automated agencies, “faceless” online content channels on TikTok and YouTube, and even car rental agencies. In one case, we discovered Naïve infrastructure supporting a TikTok channel that posted AI-generated videos of dogs and cats dancing and boxing.
“I think the fastest growing industry right now is AI automation agencies,” Dorje said. “As you know, the first business many people start is purely selling their agency to other small and medium-sized businesses (…) We also have customers who run entire car rental agencies autonomously.”
But this toolkit for running a business may be just part of the opportunity for Naïve. Using AI to automate everything sounds great, but the cost of keeping an agent running can grow astronomically as the agent calls expensive AI models, passes large amounts of context between tasks, and consumes resources while sitting idle.
Naïve says some of the new funding will be used to develop infrastructure to make the agent loop more efficient. We are building a model router that queries the most efficient model for the task, while storing and replaying data that has already been inferred. A memory system that stores and displays the business context that agents need to perform their tasks. and an orchestrator for dividing work among agents.
Notably, Naïve is also building a serverless runtime that runs agents within a lightweight JavaScript environment, rather than allocating a full virtual machine to each agent. This approach allows customers to pay primarily when agents are active, making it cheaper to deploy large numbers of agents.
Autonomous enterprise toolkits are currently in the highest demand, and inference cost optimization is one of the fastest growing sources of demand, Dorje said. “As part of running an autonomous company and running agents, that’s the biggest cost line right now, and I think what’s most in demand right now is inference and serverless agents,” he said.
He added that some aspects of the project are also attracting interest from companies, although he did not name specifics.
That could be a more valuable business than helping founders automatically set up phone numbers and corporate cards. Developers may initially use Naïve to deal with the hassle of setting up a company, but as they grow, they will focus on whether it can meaningfully reduce the recurring costs of running a large number of agents. Companies with established businesses may have some considerations on this front as well.
Naïve currently has 10 full-time employees. Dorje said proceeds from the Series A will be used to hire researchers and develop the company’s four infrastructure projects. Optimizing model routing and inference. storage layer, and governance and orchestration.
In addition to Y Combinator, Zetta, and Liquid 2, angel investors including Gokul Rajaram, Apollo.io co-founder Tim Zheng, and former HubSpot COO JD Sherman also participated in the Series A. With this latest funding, the company’s total capital raised is approximately $32 million.
If you buy through links in our articles, we may earn a small commission. This does not affect editorial independence.
