CV Stress Test

Simulate N hiring managers reading your CV. See what they think.

Paste your CV + a job description → N independent AI personas evaluate it in parallel → synthesized verdict with advance rate, suggested roles with fit score, strengths, concerns, CV adjustments, and course recommendations.

Demo

cv-stress-test.vercel.app

Support this project

This tool is free and open source. If it helped your job search:

Deploy your own (free, 5 min)

1. Fork this repo
2. Get a free API key at console.groq.com (no credit card)
3. Deploy to Vercel — add GROQ_API_KEY as environment variable
4. Share with your community

Run locally

git clone https://github.com/GerardoLucero/cv-stress-test
cd cv-stress-test
npm install
cp .env.example .env.local   # add your GROQ_API_KEY
npm run dev

API

curl -X POST https://cv-stress-test.vercel.app/api/analyze \
  -H "Content-Type: application/json" \
  -d '{
    "cv": "Your CV text here",
    "jobDescription": "Job description here",
    "n": 5
  }'

How it works

1. Generates N industry-specific hiring manager personas matched to the role
2. Each manager evaluates your CV independently (concurrency pool, respects Groq limits)
3. Synthesizes all reactions into actionable feedback:
   • Advance rate — % of managers who would move you forward
   • Suggested roles with fit score — what you can apply to now
   • CV adjustments — section-by-section specific actions
   • Course recommendations — targeted courses to close skill gaps
   • PDF report — downloadable via browser print

Tech stack

• Next.js 15 · TypeScript
• Vercel AI SDK + Groq (LLaMA 3)
• Stateless — no database required
• Deployed on Vercel

Models: llama-3.1-8b-instant (evaluations) · llama-3.3-70b-versatile (synthesis)
Provider: Groq — free tier, no credit card required

License

MIT