Building a Zero-Trust Faculty Recruitment Agent & AI Triage Pipeline with Sanity CMS & Gemini published
This is a submission for the Sanity Challenge, Path Two: Vibe-Code Something Strange What I Built Most AI recruitment chatbots on the web are toy implementations: ungrounded wrappers that hallucinate job requirements, invent benefits, and indiscriminately tell applicants, "You’d be a great fit, apply now!" In high-stakes domains like school education, ungrounded AI is a legal and regulatory…
This submission is for the Sanity Challenge, Path Two: Vibe-Code Something Strange. The project built is a zero-trust faculty recruitment agent that utilizes AI to triage applications in compliance with legal and regulatory requirements. Traditional AI recruitment chatbots often produce ungrounded outputs, such as fabricating job requirements and benefits, which present legal liabilities in education sectors like school education.
The regulations enforced by boards like CBSE and NCTE mandate specific qualifications for teacher appointments, and hiring without these can lead to disaffiliation and loss of accreditation.
The Zero-Trust Faculty Recruitment Agent & Human-in-the-Loop AI Triage Pipeline is composed of several components. The public-facing AI Career Advisor (/careers) is a conversational agent built with Google Gemini 2.0 Flash and the Vercel AI SDK, which strictly adheres to the information stored in Sanity CMS through live GROQ tool-calling. It handles candidate inquiries, verifies statutory eligibility in real-time, rejects ineligible qualifications, and suggests legal alternatives.
The automated zero-trust pre-screening (/api/career/apply) component processes cover letters and resumes upon application submission. It runs zero-assumption statutory verification, scores the match, and records missing qualifications directly into Sanity. The staff hiring & AI triage portal (/staff/careers) provides an authenticated staff dashboard for pipeline metrics, status updates, and candidate management.
The principal's executive dossier synthesizer (/api/staff/careers/generate-dossier) creates a confidential briefing for school leadership before interviews, which includes de-noised resume summaries, statutory compliance checks, and a human inquiry guide with pedagogical probe questions.
The live demo and interactive testing guide allow judges and reviewers to test the production system without local setup. The live web application can be accessed at https://school-dun-three.vercel.app, and the live careers page & AI advisor can be found at https://school-dun-three.vercel.app/careers. For staff login, the URL is https://school-dun-three.vercel.app/admin/login, with the staff ID STF-015 and password staff123.
Interactive test scenarios for judges include querying live structured content from Sanity, performing a regulatory CBSE/NCTE statutory stress-test, and accessing staff triage and the principal's executive dossier. The code for this project is available on GitHub at https://github.com/revansh1710/school.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
