vercel-labs/jev-ai-sdk-form-router
Organization repository on GitHub
Route form submissions to the right people with Jev and AI SDK.
- Forks
- 5
- Last pushed
- Updated Sep 2026
Repo stats from the GitHub API, cached Oct 2026.Homepage
Jev directory
CommunityJev Demos & Experiments
Jev x AI SDK Form Router
Vercel Labs example that routes three forms by context with Jev and AI SDK: Jev's team choice is accepted when its confidence is at least 95%, otherwise openai/gpt-6-luna-fast independently evaluates the submission and its choice becomes final.
- Stars
- 12
- Language
- TypeScript
- License
- MIT
- Reading time
- 2 min
- Published by
- Community
- Author
- Vercel Labs
- Last verified
- Oct 2026
- Languages
- TypeScript
- Platforms
- Next.jsVercel AI Gateway
- Capabilities
- Open sourceRequires API key
Highlights
- Leads route to startup, growth, enterprise, or sales teams; contact to billing, support, or general; issues to five engineering teams.
- Jev evaluates the full submission with AI SDK's experimental_evaluate and selects an allowed team/specialty combination.
- Confidence and selected-option probability are separate metrics; the threshold uses the unrounded providerMetadata.typesafe.confidence.destination.
- Provider calls have bounded timeouts and one transient retry; if both models fail the app returns a retryable error and sends no email.
- Stack: Next.js 16, React 19, AI SDK 7, Vercel AI Gateway, shadcn/ui, Tailwind CSS 4, React Email, and optional Resend delivery.
Quickstart
bash
git clone https://github.com/vercel-labs/jev-ai-sdk-form-router.git
cd jev-ai-sdk-form-router
pnpm install
cp .env.example .env.local
pnpm devWatch out
Copy reviewed against the repository README on 2026-10-15; it was truncated mid-Resend setup. Live routing requires AI Gateway access to typesafe-ai/jev and openai/gpt-6-luna-fast; without credentials the forms and sample inputs load but produce no routing results.
Questions about this entry
Short answers drawn from the linked project or page.
A Vercel Labs example application where three sample forms route submissions to the right team by context. Jev evaluates each submission with AI SDK, and openai/gpt-6-luna-fast handles uncertain or failed evaluations.
Zod validates the submission against the fields in lib/examples.ts, Jev evaluates the complete submission with AI SDK's experimental_evaluate and selects an allowed team/specialty combination, and the app accepts that choice when confidence is at least 95%.
If confidence is lower, missing, or invalid, or if Jev fails, openai/gpt-6-luna-fast independently evaluates the same submission and criteria using generateText and Output.object, and its choice becomes final.
With Node 22+ and pnpm, clone the repo, run pnpm install, copy .env.example to .env.local, set AI_GATEWAY_API_KEY or link a Vercel project for OIDC, then pnpm dev. The home page redirects to /leads.
Routing a submission makes live model calls and needs AI Gateway access to both typesafe-ai/jev and openai/gpt-6-luna-fast. You can skip that step to explore the forms and load sample inputs without generating routing results.
Email delivery is optional and uses Resend through the Vercel Marketplace. If both models fail, the app returns a retryable error and sends no email.
The destination, the deciding model, Jev statistics, model timings, and an email preview. Jev's displayed statistics stay attached to its original evaluation even when the fallback model makes the final decision.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
X: Jev lands on the Vercel AI Gateway
X: Vercel ships the AI SDK provider for Jev
YouTube: Greg Isenberg and Ryan Vogel on the businesses Jev unlocks, and a path in through the Vercel AI Gateway.
More Jev resources like this
A Next.js console whose whole design system is generated at runtime by Jev from a one-sentence brief: type 'an 18th century printed book' and every visual decision is chosen, not written, from a catalog a person authored.
Jev Demos & Experiments#community#catalog-import#madewithjev
Community
bilune/jev-designTypeScript
bilune/jev-designTypeScriptA Next.js and MUI showcase of Jev classifying Enron-Spam emails: one call asks a Noul spam question, a Choice category question and a Score urgency question at once. Includes a dataset view, a no-key example tab and a stand-alone Python script.
Jev Demos & Experiments#community#catalog-import#logicrw
Community
harsha89/jev-demoTypeScript
harsha89/jev-demoTypeScriptJevSpeak explores whether a model that cannot generate text can still talk: about 13 parallel Jev decisions per turn become a typed Semantic IR that a deterministic compiler renders in English or Chinese. Live at jevspeak.org.
Jev Demos & Experiments#community#catalog-import#logicrw
Community
mm-sheng/jevspeakTypeScript
mm-sheng/jevspeakTypeScriptAn interactive Next.js workbench for Jev's three decision primitives through OpenRouter: rank 95+ categories with one Choice pass, test binary Noul decisions with transfer curves, and grade inputs on ordinal Score rubrics.
Jev Demos & Experiments#community#catalog-import#logicrw
A BYOK workbench for practicing Jev concepts without official access: emulates State, Noul, Score and Choice via Groq, OpenAI, Anthropic, Gemini or local Ollama, or runs against native Jev with authorized credentials.
Jev Demos & Experiments#community#catalog-import#logicrw
A TypeScript starter for Jev called through the Vercel AI Gateway with the AI SDK. It wraps evaluate calls in an askJev helper and ships a support-ticket triage example using choice, score and boolean questions.
Jev Demos & Experiments#community#github#typescript
Community
1leetadmin/JEVTypeScript
1leetadmin/JEVTypeScriptFrom the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.