🧠 AI Resume Improver — Apify Actor

Upload a resume (PDF, DOCX, or paste text) → get an ATS score (0-100), missing keywords, and actionable improvements tailored to 20+ job roles.

🚀 Quick Start

Deploy to Apify

# 1. Install Apify CLI
npm install -g apify-cli

# 2. Login
apify login

# 3. Navigate to this folder
cd apify-resume-improver

# 4. Push to Apify
apify push

Run Locally

# Install dependencies
pip install -r requirements.txt

# Run with Apify CLI
apify run --input='{"resumeUrl": "https://example.com/resume.pdf", "targetRole": "Software Engineer"}'

📥 Input

Field	Type	Required	Description
resumeUrl	string	No*	Direct URL to a PDF or DOCX file
resumeText	string	No*	Paste resume text directly
targetRole	enum	Yes	Job role to match keywords against (20 roles available)
customKeywords	string	No	Comma-separated extra keywords from a specific job posting

* Provide either resumeUrl or resumeText.

Supported Roles

AI/ML Research Scientist, Backend Developer, Business Analyst, Cloud Architect, Cybersecurity Analyst, Data Analyst, Data Scientist, DevOps Engineer, Financial Analyst, Frontend Developer, Full Stack Developer, HR Manager, Machine Learning Engineer, Marketing Manager, Mobile Developer, Product Manager, Project Manager, Sales Executive, Software Engineer, UX/UI Designer

📤 Output

Results are pushed to the default dataset and stored in the key-value store as OUTPUT.

{
  "atsScore": 72,               // 0-100 overall score
  "grade": "B",                 // A / B / C / D
  "targetRole": "Software Engineer",
  "wordCount": 487,
  "bulletPoints": 12,
  "quantifiedMetrics": 5,
  "keywordMatchPercent": 62.5,

  "scoreBreakdown": {
    "keywordMatch":     { "score": 19, "max": 30 },
    "contentRelevance": { "score": 10, "max": 15 },
    "resumeSections":   { "score": 13, "max": 15 },
    "quantifiedImpact": { "score": 13, "max": 15 },
    "actionVerbs":      { "score": 7,  "max": 10 },
    "formattingLength": { "score": 7,  "max": 10 },
    "contactInfo":      { "score": 5,  "max": 5  }
  },

  "keywords": {
    "found": ["python", "react", "docker", ...],
    "missing": ["kubernetes", "system design", ...],
    "total": 45
  },

  "actionVerbs": {
    "strong": ["built", "designed", "implemented", ...],
    "weak": ["responsible for", "helped"]
  },

  "sectionsDetected": {
    "summary": true,
    "experience": true,
    "education": true,
    "skills": true,
    "certifications": false,
    "projects": true,
    ...
  },

  "contactInfo": {
    "email": "john@example.com",
    "phone": "+1 555-123-4567",
    "linkedin": "linkedin.com/in/johndoe"
  },

  "improvements": [
    {
      "category": "keywords",
      "priority": "critical",
      "message": "Keyword gap is critical (42% match)...",
      "suggestedKeywords": ["kubernetes", "system design", ...]
    },
    ...
  ]
}

📊 Scoring Rubric (100 points)

Category	Weight	What It Measures
Keyword Match	30 pts	Role-specific keywords found vs missing
Content Relevance	15 pts	TF-IDF cosine similarity to ideal profile
Resume Sections	15 pts	Essential (summary, experience, education, skills) + bonus sections
Quantified Impact	15 pts	Numbers, percentages, dollar amounts in bullets
Action Verbs	10 pts	Strong verbs vs weak passive phrases
Formatting & Length	10 pts	Word count sweet spot + bullet point usage
Contact Info	5 pts	Email, phone, LinkedIn presence

🔗 API Usage

# Run via Apify API
curl -X POST "https://api.apify.com/v2/acts/YOUR_USERNAME~ai-resume-improver/runs" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_TOKEN" \
  -d '{
    "resumeUrl": "https://example.com/resume.pdf",
    "targetRole": "Data Scientist",
    "customKeywords": "langchain, RAG, vector database"
  }'

🛠️ Tech Stack

• Python 3.11 + Apify SDK
• pdfplumber — PDF text extraction
• python-docx — DOCX parsing
• scikit-learn — TF-IDF vectorization & cosine similarity
• httpx — Async HTTP for file downloads