Overview
Maester is an AI-powered hiring panel that searches live job listings, scores them against a candidate’s resume, and generates tailored, ATS-optimized applications. Five distinct reviewer personas evaluate each role through the lenses of hiring, product, engineering, design, and recruiting—and surface disagreement instead of collapsing into generic praise.
Problem
Candidates are told to tailor every application, but job boards provide little help deciding which listings deserve that time. Noisy or stale postings waste effort, while ungrounded AI-generated applications can fabricate experience and undermine credibility.
Product approach
Live search spans Remotive, Greenhouse, Ashby, Gem, and Lever. Claude Haiku performs fast triage; Claude Sonnet runs deeper multi-agent evaluation. The tailoring engine produces ATS-safe resume and cover-letter PDFs with code-enforced rules against inventing employers, dates, experience, or metrics.
Key lessons
The working product combines live data from five job-board sources, multi-model cost optimization, local CSV tracking, opt-in email, and explicit fabrication guardrails.
Stack
Streamlit · Anthropic API · Remotive · Greenhouse · Ashby · Gem · Lever · ReportLab · Python