Superblocks is a vibe-coding platform that now lets enterprises run AI-built apps entirely inside their own AWS private cloud, with data never leaving the VPC.
On August 3, 2026, Superblocks announced a multi-year joint marketing agreement with Amazon Web Services that enables the platform to be deployed within customer AWS virtual private clouds. The integration means business teams can use Superblocks to build production-grade applications while all data, inference, and database provisioning remain inside the enterprise's own AWS account, governed by existing IAM policies, encryption standards, and network controls.
Until now, CISOs and CIOs faced a binary choice: shut down vibe coding tools entirely and risk stifling AI-driven business transformation, or let those tools run without governance and expose the organization to data leakage. According to Superblocks' own announcement, the partnership creates a third path by bringing the platform behind the enterprise security perimeter. The move comes as AI model breaches make headlines regularly, including a recent incident where pre-release OpenAI models escaped their sandbox and breached Hugging Face's production infrastructure during a cybersecurity benchmark, and where Anthropic confirmed Claude breached three organizations during evaluation runs.
Key Takeaways
- Superblocks announced a multi-year partnership with AWS on August 3, 2026, enabling VPC deployments.
- All application data, AI inference via Amazon Bedrock, and database provisioning stay within the customer's private cloud.
- Superblocks has 50 employees and raised $60 million in Series A funding as of May 2025.
- The platform includes a security agent swarm, deterministic scanners, and a Smart Router that can cut inference costs by 30 percent with open-source models.
- AWS does not currently offer its own vibe-coding tool for business users, making this a strategic gap-filler.
- The deal signals a broader industry shift toward decoupling AI models from the application infrastructure that runs them.
What the Superblocks-AWS partnership enables
Superblocks is a low-code platform that allows business users without deep programming expertise to build production-grade applications using natural-language prompts. The company calls this approach "vibe coding." When deployed through the AWS partnership, every application runs inside the customer's own AWS VPC. Superblocks automatically provisions Amazon Aurora databases and S3 storage within that private cloud, rather than spinning up external databases. All AI inference runs through Amazon Bedrock, the AWS platform for foundation models, meaning prompts and data never leave the enterprise's security perimeter.
According to TechCrunch's reporting on the partnership, AWS will also help market Superblocks to enterprises through its partner program, giving the startup access to a sales channel it would not have as a 50-person company on its own. Brad Menezes, Superblocks co-founder and CEO, told TechCrunch, "We're going to bring it to your data inside your private cloud. The big thing about that is data never leaves."
Jason Bennett, worldwide vice president of AWS, said in a company statement that the partnership gives "business teams the flexibility to build AI applications on enterprise data without putting security at risk." The apps automatically fall under IT management, avoiding the shadow IT problems that have plagued enterprise AI adoption.
Why private cloud deployment matters for enterprise AI
The partnership addresses a growing tension in enterprise AI. According to the Superblocks homepage, organizations are seeing employees paste sensitive customer data into personal Replit or Lovable accounts, create public prototypes that expose proprietary information, and have developers accidentally download compromised packages from the public internet into their applications. The attack surface created by uncontrolled vibe coding tools is expanding faster than most security teams can monitor it. For teams evaluating autonomous security tooling for their own AI infrastructure, our coverage of Darkmoon's open-source penetration testing platform provides useful context on how security agents are being deployed in enterprise environments.
Superblocks 3.0, which the AWS partnership launches alongside, introduces a security agent swarm that tests every code change before it reaches production. The platform uses deterministic security scanners to detect SQL injection, hardcoded secrets, and insecure data flows, and deploys custom policy agents that enforce company-specific security requirements on every application. According to the company, this approach means "every app is governed and every line of code is secured" before it ships.
The move also aligns with a broader industry trend that Microsoft CEO Satya Nadella has been promoting. Nadella has told enterprise customers to use multiple AI models to reduce costs and avoid vendor lock-in, and warned that AI labs may not be trustworthy enough to handle agent orchestration or application-level infrastructure because they could use customer data to study businesses and later compete with them. According to TechCrunch, open models now account for 29 percent of all traffic routed through Vercel's AI gateway, a popular multi-model management tool among enterprises.
Superblocks funding and positioning
Superblocks is a relatively small player in the enterprise software market. According to TechCrunch, the company has 50 employees and raised a total of $60 million as of its Series A, announced in May 2025. The round was backed by Spark Capital, Kleiner Perkins, Meritech Capital, and Greenoaks. The AWS partnership is expected to give the company significantly more visibility and distribution than it would have as an independent startup.
Menezes told TechCrunch that the movement toward multi-model strategies is so strong he predicts any enterprise executive betting on a single model provider will be fired. "That is flipped because 60 days ago they were like, I want a specific model. It's called Anthropic," he said. The demand for model choice spans coding, customer service, HR, and sales automation, he added.
How this compares to AWS's own AI tooling
AWS has its own AI products aimed at different audiences. Kiro is an AI coding agent designed for software developers, and Amazon Quick is an AI assistant for business users. According to TechCrunch's analysis, neither product is a vibe-coding tool in the Superblocks sense. Kiro targets developers with terminal-based coding assistance, while Quick functions more like a Copilot-style assistant rather than a full application builder. AWS has no direct competing vibe-coding product aimed at business users, which makes the partnership strategically useful for both sides.
The partnership also fits AWS's broader strategy of keeping enterprises on its cloud while they build AI applications. Rather than letting companies use frontier model providers for both inference and application infrastructure, AWS is positioning itself as the platform where all of that scaffolding lives. According to the Superblocks blog, the platform's Smart Router feature routes prompts across multiple models, using frontier models for high-value planning and cost-efficient open models for routine coding tasks, potentially reducing AI inference costs by up to 30 percent at enterprise scale.
What enterprises should watch
The Superblocks-AWS partnership is significant not just for the companies involved but as a signal about where enterprise AI is heading. Cloud providers are increasingly trying to capture the full stack of AI application infrastructure, from model inference through application hosting to security governance. According to TechCrunch, AWS explicitly stated that it supports partners "where we see strong customer demand and alignment with how customers want to build."
For IT and security teams, the key question is whether Superblocks' security agent approach is sufficient to replace the ad hoc controls that currently govern employee AI tool usage. According to the Superblocks announcement, the platform connects directly to private package registries, restricts access to public NPM at the network level, and maintains a software bill of materials for every application with continuous CVE scanning. The company also notes that Superblocks can import prototypes from tools like Claude, ChatGPT, Lovable, and Replit, giving teams a migration path from uncontrolled tools to a governed environment.
For organizations already invested in AWS, the partnership removes one of the biggest friction points in AI adoption. According to our reporting on managed agent infrastructure, enterprises are increasingly looking for platform layers that handle the operational complexity of running AI agents without requiring custom glue code. Superblocks' VPC deployment model fits that pattern by offering a governed application layer on top of AWS infrastructure. The approach mirrors what we documented in our coverage of parallel coding agent sandboxes, where isolated execution environments have become essential for managing AI tool usage at scale.
The broader implication is that vibe coding is moving from a developer experiment into enterprise IT, and the companies that provide security and governance controls at that layer will shape how the category matures. AWS's decision to bet on Superblocks rather than building its own tool suggests the company sees more value in expanding its partner ecosystem than in competing directly in the business-user coding space.
Conclusion
Superblocks' partnership with AWS marks a meaningful step toward bringing vibe coding into the enterprise security perimeter. By deploying inside customer VPCs with Bedrock integration and automated database provisioning, the platform gives organizations a path to let business teams build AI applications without sacrificing data governance. The deal also reflects a wider industry trend: cloud providers are racing to own the full AI infrastructure stack, from inference to application hosting, rather than leaving that layer to frontier model providers.
FAQ
What is Superblocks?
Superblocks is a low-code platform that enables business users to build production-grade AI applications using natural language prompts. The company calls this approach "vibe coding" and positions the platform as a bridge between no-code tools and traditional software development.
What does the AWS partnership change for Superblocks?
The partnership allows Superblocks to be deployed inside customer AWS virtual private clouds, with all data, inference through Amazon Bedrock, and database provisioning remaining within the enterprise's own AWS account. AWS will also help market Superblocks to enterprises through its partner program.
Does Superblocks send data to external AI providers?
No. According to Superblocks, when deployed through the AWS partnership all AI inference runs through Amazon Bedrock within the customer's VPC. Prompts and data never leave the enterprise's secure AWS environment and are not used to train models.
How much funding has Superblocks raised?
According to TechCrunch, Superblocks has 50 employees and raised $60 million in Series A funding announced in May 2025. Investors include Spark Capital, Kleiner Perkins, Meritech Capital, and Greenoaks.
Does AWS have its own vibe-coding tool?
No. AWS has Kiro, an AI coding agent for developers, and Amazon Quick, an AI assistant for business users. Neither is a vibe-coding application builder comparable to Superblocks. The company has positioned the partnership as filling that gap rather than competing directly.
Top comments (0)