Career-Copilot

Companies use AI to filter candidates. This gives candidates AI to choose companies.

Demo recording · compatible with asciinema players · source tape can be regenerated with VHS

---

AI-powered job search pipeline for AI coding agents. Works with GitHub Copilot CLI, Claude Code, Gemini CLI, Cursor, Windsurf, and any compatible AI tool. Evaluate offers, generate tailored CVs, scan portals, track applications, and prepare for interviews — all from your terminal.

What It Does

• Evaluate job offers with a structured A-F scoring system (6 weighted dimensions)
• Generate ATS-optimized, personalized CVs per job description
• Track your application pipeline with integrity checks
• Scan 45+ pre-configured company job boards
• Batch-process offers in parallel
• Build interview story banks (STAR+Reflection framework)
• Research companies in depth
• Draft LinkedIn outreach messages

Important: This is NOT a spray-and-pray tool. career-copilot is a filter — it helps you find the few offers worth your time out of hundreds. The system strongly recommends against applying to anything scoring below 4.0/5. Your time is valuable, and so is the recruiter's. Always review before submitting.

Heads up: the first evaluations won't be great. The system doesn't know you yet. Feed it context — your CV, your career story, your proof points, your preferences. The more you nurture it, the better it gets.

Features

Feature	Description
Auto-Pipeline	Paste a URL, get a full evaluation + PDF + tracker entry
6-Block Evaluation	Role summary, CV match, level strategy, comp research, personalization, interview prep (STAR+R)
Interview Story Bank	Accumulates STAR+Reflection stories across evaluations
ATS PDF Generation	Keyword-injected CVs with Space Grotesk + DM Sans design
Portal Scanner	45+ companies pre-configured across Greenhouse, Ashby, Lever, Wellfound
Feed Scanner	RSS/Atom job discovery with keyword filters and pipeline deduplication
Batch Processing	Parallel evaluation with task tool sub-agents
Pipeline Integrity	Automated merge, dedup, status normalization, health checks
Analytics & Negotiation	Funnel metrics, response rates, startup equity analysis, and offer negotiation workflows
Human-in-the-Loop	AI evaluates and recommends, you decide and act

Demo

The preview above is a GitHub-renderable snapshot of the shipped terminal flow. The lightweight recording lives at docs/demo.cast and can be played with an asciinema-compatible player:

asciinema play docs/demo.cast

For launch assets, use docs/assets/social-preview.svg and the checklist in docs/LAUNCH.md. For a GIF, install VHS and record the same flow from the commands in the cast: paste a job URL, generate an evaluation, create a PDF, and update the tracker.

Quick Start

Prerequisites

• An AI coding agent — any of:
  • GitHub Copilot CLI
  • Claude Code
  • Gemini CLI
  • Cursor
  • Windsurf
  • Or any AI tool that reads project instruction files
• Node.js >= 18 (for PDF generation)

Setup

Choose the path that matches how you want to use the project:

• Template repository: best for a private personal copy on GitHub before adding cv.md or config/profile.yml.
• Private fork: best if you want to pull upstream changes while keeping personal data out of public commits.
• Local clone: best for trying the project without publishing any personal configuration.

# Option A: One-liner setup
git clone https://github.com/RajjjAryan/career-copilot.git
cd career-copilot && bash setup.sh

# Option B: Manual setup
git clone https://github.com/RajjjAryan/career-copilot.git
cd career-copilot && npm install
npx playwright install chromium   # Required for PDF generation

# 2. Check setup
npm run doctor                     # Validates all prerequisites

# 3. Configure
cp config/profile.example.yml config/profile.yml  # Edit with your details
cp templates/portals.example.yml portals.yml       # Customize companies

# 4. Add your CV
# Create cv.md in the project root with your CV in markdown
# See examples/cv-example.md for reference

# 5. Start using — open your AI coding tool in this directory

For the full setup path, including privacy boundaries and AI-tool configuration, see docs/SETUP.md.

Usage

Open the project in your terminal and start your AI coding tool. The agent reads the project instruction file (INSTRUCTIONS.md) automatically. Just ask naturally:

# Evaluate a job offer
"Evaluate this job: https://jobs.lever.co/company/abc123"

# Or paste a job description directly
"Evaluate this: [paste JD text]"

# Generate a tailored CV
"Generate a PDF for this role at Google"

# Scan portals for new offers
"Scan job portals for new offers"

# Check your pipeline
"Show my application tracker"

# Compare multiple offers
"Compare my top 3 offers"

# Prep for an interview
"Prep me for my interview at Stripe for Senior Backend Engineer"

# Process pending URLs
"Process my pipeline"

# LinkedIn outreach
"Help me reach out to the hiring manager at Vercel"

How It Works

You paste a job URL or description
        │
        ▼
┌──────────────────┐
│  Archetype       │  Classifies: Backend, Frontend, AI/ML, DevOps, PM, SA
│  Detection       │
└────────┬─────────┘
         │
┌────────▼─────────┐
│  A-F Evaluation  │  Match, gaps, comp research, STAR stories
│  (reads cv.md)   │
└────────┬─────────┘
         │
    ┌────┼────┐
    ▼    ▼    ▼
 Report  PDF  Tracker
  .md   .pdf   .tsv

Project Structure

career-copilot/
├── .github/
│   ├── copilot-instructions.md   # AI agent instructions (Copilot CLI)
│   └── ISSUE_TEMPLATE/           # Bug report & feature request
├── modes/                         # Workflow definitions and localized variants
│   ├── _shared.md                # System rules, scoring, archetypes
│   ├── _profile.template.md      # User customization template
│   ├── evaluate.md               # Single offer evaluation (A-F)
│   ├── auto-pipeline.md          # Full pipeline (eval + PDF + track)
│   ├── pdf.md                    # ATS-optimized PDF generation
│   ├── compare.md                # Multi-offer comparison
│   ├── scan.md                   # Portal scanner
│   ├── feed-scan.md              # RSS/Atom feed discovery
│   ├── batch.md                  # Batch processing
│   ├── pipeline.md               # URL inbox processing
│   ├── tracker.md                # Application status dashboard
│   ├── analytics.md              # Funnel metrics and response analysis
│   ├── apply.md                  # Application form assistant
│   ├── auto-apply.md             # Human-reviewed application workflow
│   ├── contact.md                # LinkedIn outreach
│   ├── deep.md                   # Deep company research
│   ├── interview-prep.md         # Interview intelligence
│   ├── negotiate.md              # Offer negotiation prep
│   ├── equity.md                 # Startup equity analysis
│   ├── patterns.md               # Outcome and role-fit pattern analysis
│   ├── training.md               # Course/cert evaluation
│   ├── project.md                # Portfolio project evaluation
│   └── de/, es/, fr/, hi/, ja/, pt/ # Localized mode packs
├── config/
│   └── profile.example.yml       # Profile template
├── templates/
│   ├── cv-template.html          # ATS-optimized CV template
│   ├── cv-i18n.yml               # Localized CV section labels
│   ├── states.yml                # Canonical application statuses
│   ├── portals.example.yml       # Portal scanner config (45+ companies)
│   └── feeds.example.yml         # RSS/Atom feed scanner config
├── fonts/                         # Space Grotesk + DM Sans (.woff2)
├── batch/
│   └── batch-prompt.md           # Self-contained batch worker prompt
├── docs/
│   ├── ARCHITECTURE.md           # System architecture with diagrams
│   ├── SETUP.md                  # Full setup guide
│   ├── CUSTOMIZATION.md          # Customization guide
│   ├── LAUNCH.md                 # Launch and sharing checklist
│   └── assets/                   # Demo preview and social card source
├── examples/
│   ├── cv-example.md             # Sample CV
│   ├── sample-report.md          # Sample evaluation report
│   └── article-digest-example.md # Sample proof points
├── data/                          # Your application data (gitignored)
├── reports/                       # Evaluation reports (gitignored)
├── output/                        # Generated PDFs (gitignored)
├── jds/                          # Saved job descriptions (gitignored)
├── interview-prep/               # Interview prep files (gitignored)
├── INSTRUCTIONS.md               # AI agent instructions (the brain)
├── CLAUDE.md                     # Claude Code entry point
├── GEMINI.md                     # Gemini CLI entry point
├── cv.md                         # Your canonical CV (create this)
├── generate-pdf.mjs              # HTML→PDF via Playwright
├── import-cv.mjs                 # Import an existing PDF/DOCX/text CV
├── feed-scan.mjs                 # Scan configured RSS/Atom feeds
├── analytics.mjs                 # Application funnel analytics
├── analyze-patterns.mjs          # Pattern analysis across reports
├── doctor.mjs                    # Setup validation
├── verify-pipeline.mjs           # Pipeline health check
├── merge-tracker.mjs             # Merge batch tracker additions
├── dedup-tracker.mjs             # Remove duplicate entries
├── normalize-statuses.mjs        # Normalize status aliases
├── check-liveness.mjs            # Job URL liveness checker
├── cv-sync-check.mjs             # Config consistency check
├── test-all.mjs                  # Full test suite
├── CONTRIBUTING.md               # Contribution guidelines
├── LEGAL_DISCLAIMER.md           # Legal disclaimer & acceptable use
├── DATA_CONTRACT.md              # User vs system data boundary
├── LICENSE                       # MIT license
└── package.json                  # Node.js dependencies & scripts

Supported AI Tools

Tool	Instruction File	Auto-Detected
GitHub Copilot CLI	.github/copilot-instructions.md	✅
Claude Code	CLAUDE.md	✅
Cursor	.cursorrules	✅
Windsurf	.windsurfrules	✅
Gemini CLI	GEMINI.md	✅
Other tools	Read INSTRUCTIONS.md manually	—

All entry-point files reference the canonical INSTRUCTIONS.md at the repository root.

Pipeline Integrity Scripts

npm run doctor       # Setup validation
npm run verify       # Pipeline health check
npm run merge        # Merge batch tracker additions
npm run dedup        # Remove duplicate entries
npm run normalize    # Normalize status aliases
npm run sync-check   # Config consistency check
npm run liveness     # Check if job URLs are still active
npm run analytics    # Show pipeline conversion metrics
npm run feed-scan    # Scan RSS/Atom job feeds
npm run import-cv -- ./resume.pdf  # Convert a PDF resume to cv.md
npm test             # Run full test suite

Data Contract

Layer	Files	Rule
User (never auto-updated)	cv.md, config/profile.yml, modes/_profile.md, data/, reports/, output/*	Your data, your control
System (safe to update)	modes/_shared.md, modes/.md (except _profile.md), templates/, *.mjs	Improves with updates

See DATA_CONTRACT.md for full details.

Customization

The system is designed to be customized by the AI agent. Just ask:

• "Change the archetypes to data engineering roles"
• "Update my profile"
• "Add these companies to my portals"
• "Adjust the scoring weights"
• "Change the CV template design"

See docs/CUSTOMIZATION.md for details.

Disclaimer

career-copilot is a local, open-source tool — NOT a hosted service. By using this software, you acknowledge:

1. You control your data. Your CV, contact info, and personal data stay on your machine and are sent directly to the AI provider you choose. We do not collect any data.
2. You control the AI. The tool can auto-submit applications on no-login platforms (Greenhouse, Lever) and flags you afterward. For login-required platforms, it prepares everything for you to submit.
3. You comply with third-party ToS. Use this tool in accordance with the Terms of Service of career portals you interact with.
4. No guarantees. Evaluations are recommendations, not truth. AI models may hallucinate.

See LEGAL_DISCLAIMER.md for full details.

Contributing

Contributions welcome! See CONTRIBUTING.md for guidelines.

Check out the good first issues to get started.

Contributors

Thanks to these wonderful people for their contributions!

Want to contribute? Check out our Contributing Guide and browse good first issues.

Star History

License

MIT — see LICENSE.