Superblocks is getting a larger enterprise channel for vibe coding through Amazon Web Services. The startup announced a multi-year joint marketing agreement with AWS that allows its tool to be embedded inside the private clouds of AWS customers.
For companies already running on AWS, the arrangement points to a specific version of business-user AI app development: one where new apps sit under existing IT management, security controls, and cloud infrastructure instead of operating as external tools.
What The AWS Agreement Changes
Under the agreement, an enterprise AWS customer that subscribes to Superblocks can offer vibe coding to business users from inside its own private cloud. The central promise is that apps built this way do not send data or information externally to model providers or databases.
That is a meaningful distinction for companies that want AI-assisted app creation but do not want business data moving outside their controlled environment. In this setup, apps spin up Amazon Aurora databases within the company’s private cloud. They do not create external Supabase databases, which the source describes as the vibe-coding database of choice.
The apps also integrate with Amazon Bedrock, AWS’s AI app development, AI gateway, and inference platform. The practical result is that applications created through Superblocks can fall under IT’s normal management and security rather than becoming rogue applications that operate outside approved systems.
“We’re going to bring it to your data inside your private cloud,” Superblocks co-founder and CEO Brad Menezes tells TechCrunch of vibe coding. “The big thing about that is data never leaves. … It’s their AWS account and basically secure with all of the auditing, all of the encryption, all of the network controls.”
Why Enterprises Care About Control
Vibe coding promises to let non-developer business users create applications more directly. For enterprises, however, the attraction of faster app creation is tied to a familiar concern: how to keep data, databases, model access, auditing, encryption, and network controls inside approved boundaries.
The AWS and Superblocks arrangement speaks directly to that concern. Instead of asking companies to accept an external app-building stack, it brings the tool into the AWS account the enterprise already uses. That makes the offering less about experimentation at the edge and more about governed deployment inside the company’s private cloud.
AWS will also help sell Superblocks to enterprises, as it does for many Marketplace partners. An AWS spokesperson tells TechCrunch, “We support partners where we see strong customer demand and alignment with how customers want to build.”
For Superblocks, the support could be significant. The company has 50 employees and had raised a total of $60 million as of its Series A, announced in May 2025. Its backers include Spark Capital, Kleiner Perkins, Meritech Capital, and Greenoaks.
A Gap In AWS’s Own AI Lineup
The agreement also matters because AWS does not yet have its own vibe-coding agent aimed at business users. AWS has Kiro, an AI coding agent aimed at developers, but the source distinguishes it from a vibe coder.
Amazon also has Quick, an AI assistant for business users. But the source describes Quick as closer to a Claude Cowork or Microsoft Copilot than to Lovable or Replit.
That makes Superblocks a useful partner for AWS in a category it does not directly fill with its own product. It gives AWS a way to support business-user app creation while keeping the experience anchored in AWS infrastructure, Amazon Aurora, and Amazon Bedrock.
The Bigger Cloud Strategy
The Superblocks agreement fits a broader pattern described in the source: hyperscaler cloud providers want enterprise customers to separate AI models from the surrounding systems needed to run enterprise AI. Those surrounding systems include AI harnesses, also described as agentic apps, along with AI orchestration, security tools, and related infrastructure.
The cloud providers want enterprises to buy that scaffolding from them, not from the frontier providers. In this framing, the model is only one part of the enterprise AI stack. The cloud platform wants to control the layer where applications are governed, routed, secured, and connected to business data.
Microsoft CEO Satya Nadella has recently been delivering a similar message to enterprise customers, according to the source. He has urged companies to use multiple models to reduce costs and avoid lock-in. He has also argued that AI labs are not trustworthy enough for agent orchestration or app-level harnesses because they may use business data to study a company and later compete with it.
Enterprises may already be moving in that direction. Menezes says the conversation has shifted from customers wanting one specific model to wanting model choice. He says that “60 days ago they were like, I want a specific model. It’s called Anthropic.”
Multi-Model AI Becomes The Enterprise Default
The source points to one example of this shift: open models accounted for 29% of all traffic routed through Vercel’s AI gateway last month. The article describes Vercel’s AI gateway as a popular tool among enterprises for managing multi-model AI use.
If enterprises use multiple models, then the rest of the AI stack cannot be tightly bound to only one provider. That applies to coding, but also to other business functions where companies want AI support.
“Having a multi-model strategy across big frontier labs, OpenAI, Anthropic, and open source — and I’d say Chinese open source right now, but also U.S. open source is now starting to come up. It’s a must-have for the CIO,” he says.
Menezes adds that companies want model choice for coding as well as customer service, HR, and sales automation. He goes further, predicting that “any enterprise that is betting on a single model provider, that executive will be fired.”
Seen that way, AWS helping Superblocks is not just a startup distribution story. It suggests that vibe coding for business users may be entering the same enterprise cloud pattern already forming around AI coding agents for developers: keep the apps close to company data, keep model choice open, and keep governance inside the cloud account.