
Vercel
Vercel Labs is the experimental GitHub organization of Vercel Inc., the San Francisco developer-platform company founded by Guillermo Rauch in 2015. Vercel created Next.js — the widely adopted open-source React framework — and the Vercel AI SDK, and uses Labs as a public sandbox for prototypes and demos that push the frontier of its platform before features graduate to the main product.
Notable people
People map
Current peopleNotable public-facing people we cover. Grouped by public roles, not reporting lines; this is not a complete staff directory.
Leadership
- Malte Ubl
Chief Technology Officer
- Guillermo Rauch
Founder and CEO
Product & engineering
- Joshua Wolk
Design Engineer · Jun 2025 – Present
- Sahaj Jain
Software Engineer
- Anthony Fu
Design Engineer
- shadcn
Design Engineer, AI team
Former people
- Emil Kowalski
Design engineer · Former
Tools
- Vercel Agent Skills
github.comVercel's official agent skills for auditing deployments and applying React and Next.js practices.
32,935 GitHub starsOriginal source - Vercel MCPvercel.com
Hosted MCP server giving agents access to Vercel projects, deployments, and build logs to investigate failed builds.
freeOriginal source - Next.jsnextjs.org
Full-stack React framework with file-based routing, server components, server actions, and an edge runtime.
- Vercel AI Gatewayvercel.com
Single endpoint and SDK routing requests to OpenAI, Anthropic, Google, and other AI providers, with caching built in.
paidOriginal source - Vercel Analyticsvercel.com
Web analytics built into Vercel that tracks Core Web Vitals without cookies.
paidOriginal source - Vercel Blobvercel.com
Managed object storage integrated with Vercel projects, with an SDK for uploads from Edge and Serverless Functions.
paidOriginal source - Open Plugin Spec
github.comMinimal standard for packaging AI agent extensions into distributable plugins.
- React View Transitions Demoreact-view-transitions-demo.labs.vercel.dev
Next.js gallery app demonstrating React View Transitions with shared element morphs and Suspense reveals.
- npm.betnpm.bet
Visualizes npm download trends across packages to show which libraries are gaining ground.
- Sandcastlesandcastle-os.vercel.app
Proof-of-concept web-based Linux desktop environment that runs on Vercel Sandbox.
- Call Summary Agent with Sandboxcall-summary-agent-demo.labs.vercel.dev
Demo agent that summarizes calls and executes follow-up code inside Vercel Sandbox.
- Claude Managed Agents Starterclaude-managed-agents.labs.vercel.dev
Open-source Next.js template for building Claude Managed Agents that deploy on Vercel.
- Nitro iMessage Agent Templatenitro-imessage-agent.vercel.app
Template for a durable iMessage AI agent with retryable LLM and tool replies, deployed on Vercel.
- QuickJS WASIquickjs-wasi.labs.vercel.dev
JavaScript runtime running QuickJS-NG in WebAssembly with snapshots of full VM state, pending promises included.
- Vercel GitHub Toolsgithub-tools.com
Wraps GitHub's API as ready-made Vercel AI SDK tools and agents that plug into any agent workflow.
- Vercel AI SDK for Pythonai-python.dev
Vercel Labs library that brings the JavaScript AI SDK's primitives to Python for building AI applications.
- Deploy OpenClaw on Vercel
github.comTemplate and tooling for deploying OpenClaw on Vercel, built with Next.js and TypeScript.
- AI SDK Multi-Step Previewai-sdk-preview-multi-steps.vercel.app
Preview demo of building multi-step agent workflows with the Vercel AI SDK and Next.js.
- Autoship
github.comCLI that automates changeset-based releases, bumping versions, writing changelogs, and publishing packages on merge.
- Markdown Sanitizersvercel.com
TypeScript sanitizers that clean AI-agent markdown output to defend rendered responses against prompt injection.
- AI SDK Slackbotai-sdk.dev
Reference Slack agent built on the Vercel AI SDK that answers questions and runs tools inside Slack.
- MCP to AI SDK
github.comAdapter that makes any MCP server's tools usable from the Vercel AI SDK inside an agent.
- AI SDK Gateway Demoai-sdk-gateway-demo.labs.vercel.dev
Next.js demo showing how to route requests between LLM providers using the Vercel AI Gateway.
- Semantic Image Searchsemantic-image-search.vercel.app
Reference build for searching images by what they show rather than filename, using embeddings and the Vercel AI SDK.
Writing, research & talks
- ai ai@7.0.153
github.comAI SDK 7.0.153 fixes non-streaming Responses API structured output for OpenAI and Azure. Providers can now supply separate text for parsing, so the last final_answer is used while content, annotations, history and usage stay intact.
Vercel explains how its Agent investigates under read-only access, then proposes a plan whose extra scopes are approved as a single decision, avoiding both broad standing permissions and approval-click fatigue.
Vercel Connect adds a Managed API Key method for Stripe connectors, creating a separate permission-limited key per agent or task, stored by Vercel and rotatable without redeploying, with CLI support.
- ai @ai-sdk/tool-web-fetch@1.0.0
github.comVercel's AI SDK published @ai-sdk/tool-web-fetch 1.0.0, its first-party web fetch tool for agents. The release note is brief, so details of its behavior and limits are not stated in the release itself.
- Send feedback from the Vercel CLIvercel.com
The Vercel CLI gained a feedback command for sending bugs and feature requests from the terminal. Agents can attach structured key=value context such as model, harness, task, expected and actual behavior, or pipe in notes from a file.
- ai ai@7.0.150
github.comAI SDK 7.0.150 adds experimental_streamSpeech, which streams audio from text deltas using speech models that implement the new streaming extension, with error handling, abort support and a mock model for tests.
- ai ai@7.0.149
github.comAI SDK 7.0.149 adds an interrupted finish reason for streams ending without a completion signal, enforces toolChoice 'none' before local execution, and processes only the first approval response per tool call.
- ai ai@6.0.313
github.comAI SDK 6.0.313 processes only the first approval response per tool call, removes orphaned reasoning when pruning tool calls, and keeps repaired tool names in streamed UI message parts.
- Death to long-lived static credentialsvercel.com
Vercel argues machine-to-machine auth still resembles pre-HTTPS web, with static keys the default. It introduces OIDC UI, a shadcn registry for trust-policy forms and token verification, widens OIDC across its platform, and sets new auth requirements for Marketplace providers.
Vercel CLI adds a temporary deploy mode that needs no account. It builds locally, returns a shareable tmp.now URL that expires after an hour, and gives a private claim link to keep the project. It works for whole projects or a single file.
Vercel's Token Pass subscription now includes DeepSeek V4.1 Flash alongside Kimi K3 and GLM 5.3. The $30/month Pro-team plan grants $80 in credit through AI Gateway and works with agents such as Pi, Cline and fx.
- Making every harness durablevercel.com
Vercel explains why cloud-hosted coding agent harnesses need durable execution, since a plain VM does not record completed model and tool calls or resume interrupted turns. It introduces eve and fx and argues for running non-sandbox work in functions.
Vercel AI Gateway logs now include a Sessions view grouping requests by session, with cost, models, providers, and cache hit rate per coding agent. Session affinity pins turns to one provider for cache reuse, using a sessionId or x-session-id header or a conversation fingerprint.
- Vercel Daemons is now in public betavercel.com
Vercel Daemons enters public beta, running containerized long-lived workers that poll queues or task systems. Vercel builds the image, keeps the configured replica count per region and restarts crashed instances.
- Introducing Token Passvercel.com
Vercel opens early access to Token Pass for Pro teams: $80 monthly credit for GLM 5.3 and Kimi K3 at $30, routed through AI Gateway with zero data retention, usable from coding agents or the AI SDK.
Vercel announces general availability of Python support: CPython on Linux, automatic framework detection, Fluid compute, and idiomatic SDKs for storage, queues, durable workflows and AI models, with a shared frontend and backend deploy workflow.
Vercel's October AI Gateway index reports open-weight models at 71.5% of tokens and 23% of spend, spend up about 13x since December, and nearly a third of teams switching primary lab in September. Data reflects Vercel's customer base only.
Vercel Connect adds public beta MCP Events: apps and agents subscribe to webhook events from MCP servers, with automatic renewal, idempotent subscriptions and SDK verification of authenticated deliveries.
Vercel Connect adds Enterprise-Managed Authorization in public beta, supporting Cross-App Access for MCP. Admins control agent access via the identity provider and Connect handles token exchange and renewal.
Vercel reports the outcome of its sandbox escape challenge: 1,285 reports, $242,000 in bounties, and a Linux kernel networking zero-day that let a microVM guest leak host memory, crash the host or execute code. No customer data access was found.
- Amazon S3 joins the Vercel Marketplacevercel.com
Vercel adds Amazon S3 to its Marketplace and v0. Resources are created from the dashboard, connection details are injected as environment variables, and access uses short-lived OIDC credentials rather than stored AWS keys.
The Vercel CLI now lets coding agents search, price-check, and initiate domain purchases using a bundled skill. Non-interactive buys return a machine-readable error so a human must confirm price, auto-renew, and contact details.
Vercel Sandbox creation now waits for snapshot restoration, so a returned sandbox is immediately ready and startup errors reject at creation. Creation is about 50 ms slower at p50, with no SDK change needed.
Vercel's AI Gateway beta now escalates decision requests to a fallback model when the primary answer trips a confidence condition, such as a score below a threshold. Conditions can be combined, and triggered fallbacks bill both stages.








