I treated my CV like a data product-evidence.json,MCP endpoint,llms.txt
Evidence-mapped CV beats PDF for AI recruiter parsing, but applies only to ATS that read these formats.
Parse a CV into a structured Turtle RDF knowledge graph using the ResumeRDF ontology via a password-protected Streamlit web app or a standalone Python script.
AI wrapper parsing CVs into existing ResumeRDF ontology — solved category.
HR teams, recruiters, developers building CV management systems
Affinda · Sovren · RChilli
Evidence-mapped CV beats PDF for AI recruiter parsing, but applies only to ATS that read these formats.
CV optimizer using LLMs when Jobscan and Resume Worded already exist.
Local-first ATS checker when Jobscan and ResumeWorded already dominate this space.
This is someone treating a CV as structured data rather than a PDF: resume.json, evidence.json, availability.json, agent-card.json and a curated llms.txt are all exposed plus schema.org JSON-LD. Nice touches include GitHub Actions that validate links and push an IndexNow update on every commit — practical engineering to get content noticed by crawlers and agents. It’s a focused, well-implemented experiment, but its usefulness depends on broader adoption or tooling that consumes these bespoke conventions.
System prompt wrapper for CV review; dozens of resume analysis tools already exist.
This is a practical playbook: the repo bundles resume.json, evidence.json, availability.json, an agent‑card and an llms.txt plus CI checks and IndexNow pushes so your CV is both human- and agent-discoverable. Clever bits: automated sitemap/index pushes, link-checking Actions, and explicit A2A‑style metadata (agent‑card.json) — that’s not something you see on most personal sites. What’s missing for wider credibility are outcome metrics and external verification (recruiter-facing analytics, attestations, or an A/B test showing improved contacts), and a clearer signal-to-noise story for what recruiters should actually consume first.