AWS and Superblocks Partner to Bring Secure Vibe Coding to Enterprise Clouds

Superblocks has signed a multiyear joint marketing agreement with Amazon Web Services (AWS) that will allow enterprises to deploy the startup’s vibe-coding platform directly inside their private AWS cloud environments. The partnership enables businesses to build AI-generated applications without sending sensitive data outside their own cloud infrastructure.

Under the agreement, applications created with Superblocks will remain inside a customer’s AWS environment instead of relying on external services. Rather than creating databases through third-party providers such as Supabase, the applications will automatically deploy Amazon Aurora databases within the customer’s private cloud. The platform also integrates with Amazon Bedrock, AWS’s managed service for building and running AI applications.

According to Superblocks co-founder and CEO Brad Menezes, the approach is designed to ensure enterprise data never leaves the customer’s AWS account while remaining protected through existing security controls, including auditing, encryption and network management.

AWS will also support Superblocks through its Marketplace partner program by helping introduce the platform to enterprise customers. An AWS spokesperson said the company supports partners that align with customer demand and preferred approaches for building AI applications.

Enterprise AI shifts toward private clouds

The agreement also highlightsAWS’ss broader enterprise AI strategy. While Amazon offers Kiro for software developers and the Quick AI assistant for business users, it does not currently provide its own vibe-coding platform designed specifically for non-technical enterprise employees. The partnership gives AWS customers access to that capability while keeping governance and security under IT control.

The deal represents an important milestone for Superblocks, which has about 50 employees and has raised $60 million, including its Series A financing announced in May 2025 with backing from Spark Capital, Kleiner Perkins, Meritech Capital and Greenoaks.

Beyond the startup itself, the partnership reflects a wider trend among major cloud providers. Rather than encouraging enterprises to depend entirely on a single AI model provider, cloud companies are increasingly positioning their platforms as the secure layer for AI orchestration, application development and governance while allowing customers to choose among multiple AI models.

Microsoft Chief Executive Satya Nadella has recently promoted a similar strategy, encouraging enterprises to adopt multiple AI models to reduce costs and avoid vendor lock-in. He has also argued that companies should avoid relying on frontier AI labs for application orchestration because those providers could potentially gain insight into customer business operations.

Multi-model AI becomes a priority

Menezes said enterprise buying patterns have changed significantly in recent months. Instead of requesting access to a single AI provider, organisations increasingly want flexibility to use multiple frontier and open-source models across coding, customer service, human resources and sales automation.

Open-source models accounted for 29% of traffic routed through Vercel’s AI gateway last month, illustrating growing enterprise adoption of multi-model AI strategies. Menezes argued that supporting several AI models has become essential for chief information officers, predicting that organisations relying on only one provider could face increasing competitive pressure.

As enterprises expand their use of AI, partnerships such as the one between AWS and Superblocks suggest that secure, private-cloud deployment of vibe coding may become the next phase of enterprise AI adoption, following the rapid rollout of AI coding assistants for professional software developers.