Vibe coding startup Superblocks has signed a multi-year joint marketing agreement with Amazon Web Services (AWS) that allows its tooling to be embedded within AWS customers' private clouds. Under the deal, an enterprise using AWS and subscribed to Superblocks can provide vibe coding to business users without sending data or information externally to model providers or third-party databases.
How it works
Superblocks-built apps will spin up Amazon Aurora databases inside the customer's own AWS account, so data stays within the company’s private cloud rather than creating external databases such as Supabase. The apps will also integrate with Amazon Bedrock, AWS's AI app development and inference gateway platform. As a result, these applications fall under IT management and security controls rather than operating as unsanctioned or shadow apps.
“We’re going to bring it to your data inside your private cloud,” Brad Menezes, Superblocks co-founder and CEO, told TechCrunch. “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.”
AWS's role in go-to-market
AWS will support selling Superblocks to enterprises similarly to how it assists other Marketplace partners. An AWS spokesperson told TechCrunch: “We support partners where we see strong customer demand and alignment with how customers want to build.”
AWS does not currently offer a vibe-coding agent aimed specifically at business users. The company has Kiro, an AI coding agent targeted at developers, and Quick, an AI assistant for business users, but those are positioned differently from the vibe-coding experience Superblocks offers.
Superblocks' size and funding
The agreement is a potential boost for Superblocks, which has about 50 employees. The company announced a total of $60 million in funding as of its Series A round, disclosed in May 2025, with investors including Spark Capital, Kleiner Perkins, Meritech Capital, and Greenoaks.
Broader industry context: multi-model strategies and cloud control
The deal also reflects a broader trend in which hyperscale cloud providers encourage enterprise customers to separate AI models from the other scaffolding required to run enterprise AI and to run that scaffolding on their clouds. Cloud providers want enterprises to source AI harnesses (agentic apps), orchestration, security tools and similar capabilities from them rather than from frontier model providers.
Microsoft CEO Satya Nadella has recently emphasized a similar message: enterprises should use multiple models to reduce costs and avoid vendor lock-in, and be cautious about relying solely on AI labs for orchestration or app-level harnesses because those labs could use business data and potentially compete later.
Enterprises have already started adopting multi-model approaches. Brad Menezes noted that preferences have shifted — where some customers previously wanted a single model such as Anthropic, they now pursue multi-model setups. For example, open models represented 29% of traffic routed through Vercel’s AI gateway last month, a sign that organizations are actively routing requests across open-source and lab models.
Menezes argued that multi-model strategies are essential for CIOs, covering use cases from coding to customer service, HR and sales automation. He warned that executives who bet on a single model provider risk negative consequences: “Any enterprise that is betting on a single model provider, that executive will be fired.”
Implications for enterprise AI
The Superblocks–AWS collaboration shows that cloud providers are moving vibe coding for business users into private, secure clouds, following an earlier wave of AI coding agents for enterprise developers. In AWS’s words, “It’s an emerging category with real momentum, and exactly the kind of innovation we support.”
By keeping databases, models, logging, encryption and audit controls inside the customer’s AWS account, enterprises can maintain IT oversight and compliance while offering AI-driven tooling to business users.
