Why does this project use Jev instead of an LLM for screening?

Jev is an evaluation model that returns choices, scores and calibrated probabilities for about $0.042 per million input tokens with output tokens free, making it cheap enough to triage every plausible posting daily.

What does the page-brain endpoint do?

Mid-application, Rudy calls secret-gated modes such as knock-out and next-action so ambiguous steps are judged by Jev before the browser fills heavy ATS forms, with every call logged to Neon.

How big is the funnel, and how much survives each stage?

The funnel starts with about 1.46 million ingested postings from the feed. Deterministic code keeps 5.6%, Jev title-only triage over title, company and location keeps 10.3% of that, and ATS describe APIs on the survivors keep 65% since the rest have expired or are unsupported.

How much does the cheap first pass save?

On 600 postings, title-only Jev triage cut token spend 41% and agreed with the full-profile call 96.5% of the time. Stage 0 deliberately handles only hard facts such as location and the recruiter flag, and leaves seniority alone because the feed's keyword skill_level tag is unreliable.

What are the two separate Jev uses?

Two, and the README insists you not mix them up. This app does title triage and full-text scoring through the AI Gateway; Rudy, the production apply side, runs a sparse page-brain that scans for knock-outs before a heavy ATS fill. The page-brain exists to catch disqualifiers, not to make filling faster.

Where does my personal data live?

In the git-ignored profile/ directory and in Vercel environment variables, never in code. The repository is public, and the stack is a Next.js dashboard with a daily Vercel cron, a Neon database and a webhook into Grok Bot.

Which job platforms can it fetch descriptions from?

Greenhouse, Lever and Ashby whole boards cached per company, plus Workday fetched per posting since no board endpoint carries text; BambooHR, iCIMS and Paylocity are not wired up, so postings there move to an unsupported stage rather than staying pending.

CommunityJev Integrations & Adapters

Grokbot Jev Jobs

A Next.js dashboard and daily Vercel cron that scores public job postings against one resume with Jev through the Vercel AI Gateway, then applies through a Grok Bot webhook that fills ATS forms and writes status back.

See highlights

Open github.com

Stars: 0
Language: TypeScript
Published by: Community
Last verified: Oct 2026

mcgalleg/grokbot-jev-jobs

User repository on GitHub

View on GitHub

Scores public job postings against my resume using TypeSafe's jev via the Vercel AI Gateway. Daily Vercel cron.

Forks: 0
Last pushed: Updated Sep 2026

Repo stats from the GitHub API, cached Oct 2026.Homepage

About this project

Scores public job postings against a resume using TypeSafe AI's Jev through the Vercel AI Gateway, then applies through a dashboard Apply button that webhooks Grok Bot (Resume Rudy) to fill ATS forms and write status back.

Quoted from the project's README. Read the full README.

Tagscommunity job-search vercel nextjs

Works forJev for Scoring Jev for Ranking Jev for Classification Jev for Decision gating

Used inJev for HR & Recruiting

Languages: TypeScript
Platforms: Vercel AI Gateway Next.js API
Capabilities: Requires API key

Highlights

Funnel starts with about 1.46M ingested postings; title-only Jev triage keeps 10.3%, then full-text scoring asks fourteen atomic questions.
On 600 postings, title-only triage cut tokens 41% and agreed with the full-profile call 96.5% of the time.
A secret-gated POST /api/rudy/page-brain endpoint answers knock-out, next-action, field-map and outcome modes mid-application.
Personal data stays in a git-ignored profile/ directory and Vercel environment variables, never in code.

Watch out

The repository publishes no license file, so reuse rights are unclear, and the deployment depends on private env vars plus a personal resume profile. Jev access goes through the Vercel AI Gateway.

Questions about this entry

Short answers drawn from the linked project or page.

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

Inbox Zero

12.4k

Open-source AI email assistant (Next.js, Postgres, 12k+ stars) with a decision-model layer that can run on openrouter:typesafe/jev-1.13: Jev-backed questions cover rule selection, cold-email detection, thread status, recurring patterns, unsubscribe pages, and sender categorization, each with evals.

Jev Integrations & Adapters#community#email#gmail

Community

elie222/inbox-zeroTypeScript

Dub

24.9k

Open-source link attribution platform that calls typesafe-ai/jev in malicious-link-check.ts before a short link is created, so the URL is gated by a typed Jev verdict rather than a static blocklist. Next.js and TypeScript with an open-core license.

Jev Integrations & Adapters#community#link-shortener#attribution

Community

dubinc/dubTypeScript

Kilo Code Auto-Routing

Merged PR in the Kilo Code cloud monorepo that replaces its LLM auto-routing classifier with one OpenRouter /systemone Choice request to typesafe/jev-1.13; the PR reports 0.93-0.94 route accuracy on 72 golden cases at roughly $0.00028 per call, with a 250 ms transport-error retry.

Jev Integrations & Adapters#community#routing#coding-agents

Community

Kilo-Org/cloudTypeScript

Jev-Mail

CLI-installed Gmail classifier that runs 24/7 on user-owned Google Apps Script: the worker asks Jev for three typed judgments per thread (requires_action, is_important, bucket), applies labels, stars, and optional archiving, and routes ambiguous or low-confidence mail to a Review label.

Jev Integrations & Adapters#community#gmail#email

Community

vynnlee/jev-mailTypeScript

Ad Spend Guard

A pre-bid spend guard that scores a placement's expected relevance and flags whether it matches the campaign brief, letting deterministic code hold budget back before the auction instead of gating the creative itself.

Jev Integrations & Adapters#community#typescript#advertising

Community

WorldMonitor

88.2k

Real-time geopolitical intelligence dashboard that uses TypeSafe Jev questions to score news headline severity into five threat levels and categorize events across 14 conflict, cyber and infrastructure domains, alongside news aggregation and a Country Instability Index.

Jev Integrations & Adapters#community#osint#geopolitics

Community

koala73/worldmonitorTypeScript

Back to all resources

From the Jev community

Posts from builders shipping with Jev right now.

Follow @typesafeai