Vercel improves developer productivity while delivering a better experience

“When I started Vercel, I saw the opportunity in creating tooling and cloud infrastructure to make the Web faster, with a focus on developer experience. It's more important than ever for ecommerce brands to deliver a fast and personalized user experience. If a page is slow, or a product shows ‘sold out’ when it isn’t, your company can lose sales or damage relationships with shoppers. We’re committed to helping companies build fast and personalized ecommerce experiences,” said Rauch

Vercel saw an opportunity to enable developers to build dynamic personalized web pages by embracing open source software, AWS infrastructure, and AI. Vercel v0 produces code using open-source technologies like React, Tailwind CSS, and Shadcn UI. A developer describes the desired functionality to v0 using natural language, which then generates the code, and the developer iteratively refines it through further conversation. Developers can deploy their application with v0 to Vercel AI Cloud or deploy the application code on the service of their choosing, such as Amazon Elastic Compute Cloud (Amazon EC2), Amazon Elastic Container Service (Amazon ECS), or Amazon Elastic Kubernetes Service (Amazon EKS). Generative UI tools such as v0 are on the frontier of democratizing design, helping non-technical and technical developers translate natural language into code with unprecedented ease.

Vercel v0 offers AWS databases, such as Amazon Aurora and Amazon DynamoDB, delivering high performance and availability for OLTP workloads and building new applications. These AWS databases, trusted by developers for over a decade, are natively integrated and available in the Vercel Marketplace. In just a few seconds, developers can connect to an Aurora PostgreSQL, Aurora DSQL, or DynamoDB database making it faster to get started. Once ready, developers can just as easily deploy into production while benefitting from the security, reliability, scalability, and operational excellence synonymous with AWS.

Vercel built a chat interface that allows developers to interact with the v0 AI assistant. v0 can output advanced Blocks that can be previewed in the chat and then copied and pasted into a project or installed via the shadcn CLI. Using Blocks, v0 can write and render Svelte, Vue, React, and HTML with CSS. Anyone with an idea can use v0 by writing prompts in their preferred language and v0 will translate them into code. Product managers create prototypes, marketing teams build landing pages, and founders ship MVPs.

Vercel v0 is built on top of AI SDK to help developers build AI-powered applications and agents. Vercel’s AI SDK incorporates two distinct types of messages:

• UIMessage: The source of truth for the application state, containing all messages, metadata, tool results, and more. Vercel recommends using UIMessages for persistence so that users can always restore the correct user-facing chat history.
• ModelMessage: A streamlined representation optimized for sending instructions to language models.

Developers can customize the UIMessage to create their own type with the exact shape of their data, tools, and metadata, that is tailored to an application. All frameworks get the same powerful features: custom message types for the application's specific needs, data parts for streaming arbitrary typed data, fully typed tool invocations with automatic input streaming, and type-safe message metadata. From server to the client, every piece of data, tool call, and metadata is fully typed. Building reliable AI agents requires precise control over execution flow and context. Vercel AI SDK contains primitives that give developers complete control over how their agents run and the context and tools they have at each step. Before each step executes, developers can adjust the LLM, switching between models based on task complexity. The Vercel AI SDK integrates with Amazon Bedrock through a dedicated provider package, @ai-sdk/amazon-bedrock. The Amazon Bedrock provider for the AI SDK contains language model support for the Amazon Bedrock APIs. This integration allows developers to leverage the powerful foundation models and generative AI capabilities offered by Amazon Bedrock within applications built using Vercel’s AI SDK. AI SDK represents the next evolution of AI libraries for the web.

Vercel's CDN is a globally distributed platform that stores content near end users and runs compute in regions close to their data. When deploying an app on Vercel, Vercel’s CDN is used by default. Developer requests hit the cloud and the end user. AWS Lambda is a serverless compute service that lives close to globally distributed servers in cloud data centers, whereas Vercel Edge Functions live right next to the users. Together they become a single fabric that moves logic close to data when it matters, and close to end users when speed matters more. “Functions from Vercel run lightweight code on the global CDN layer, executing within milliseconds of each request,” said Rauch.

The services are complementary — AWS Lambda is for deep processing or sensitive workflows, and Vercel Edge Functions are for the fast, transactional bits. AWS Lambda with Vercel Edge Functions delivers global response times under one-tenth of a second for most endpoints, centralized access control enforced through AWS IAM and OIDC, observability through consolidated structured logs, and a consistent security posture across cloud edges and private networks.

Coming soon, Vercel will integrate with Amazon CloudFront and Amazon Bedrock AgentCore to power modern applications. CloudFront strengthens Vercel’s CDN for global performance, while Amazon Bedrock enhances the AI Gateway for scalable, cost-efficient AI inference. Together they advance the goals of both companies to be the leading application provider for everything from static content to dynamic applications and AI workloads.

The world of vibe coding is transformative for organizations. Vercel builds agentic AI products that help developers work faster and improve their experience, as they write code once, deploy globally, and see instant feedback. The joint solution with AWS improves velocity and drastically cuts wait times. Vercel’s strong developer ecosystem presence and deep AWS integrations, particularly with Amazon Bedrock and AWS Lambda positions it as a leader in advancing AI application development.

“Through our partnership with AWS, we've been able to scale v0's inference on Amazon Bedrock very successfully to meet the needs of our user base,” said Rauch. Since launch, Vercel has seen 25 million UIs generated using v0, and customers are experiencing up to a 95% reduction in page load times. Developers are also experiencing up to 24x faster builds. “Amazon Bedrock has been influential throughout that entire process, serving up all those requests. And knowing that Amazon has that capacity ahead of time and can grow and scale with us has been tremendous.”