Resume Analyzer

Analyze resumes against job descriptions and generate optimized versions with detailed annotations.

When to Use

User provides a resume file (.docx) and wants analysis
User provides both resume and JD for matching analysis
User wants ATS optimization suggestions
User wants quantified achievements and stronger bullet points
User wants an optimized version with explanations

Workflow

Step 1: Load & Parse Resume

Use scripts/parse_resume.py to extract content from .docx:

python scripts/parse_resume.py <input.docx> --output parsed_resume.json

Step 2: Five-Dimension Analysis

Run analysis script with JD (if provided):

python scripts/analyze_resume.py parsed_resume.json [--jd job_description.txt] --output analysis_report.json

The analysis covers:

JD Match Score (0-100): Keyword overlap, skills alignment
Quantification Score (0-100): Presence of metrics, numbers, percentages
Structure Logic (0-100): Section order, readability, hierarchy
Language Professionalism (0-100): Action verbs, clarity, conciseness
ATS-Friendliness (0-100): Format, keywords, standard sections

Step 3: Interactive Q&A

Present the 5-dimension report and ask follow-up questions:

Questions to ask (user can select or type):

Which role at [Company X] had the biggest impact? What were the measurable results?
Any specific project with quantifiable outcomes (revenue, users, efficiency)?
Tools/technologies used that aren't mentioned?
Any awards, recognition, or leadership experiences to highlight?
Education details: GPA, relevant coursework, projects?

Store answers in supplemental_data.json.

Step 4: Generate Optimized Version

python scripts/generate_optimized.py \
  parsed_resume.json \
  analysis_report.json \
  supplemental_data.json \
  --output optimized_resume.docx \
  --backup original_backup.docx

Output files:

original_backup.docx: Clean copy of original
optimized_resume.docx: Optimized version with Word comments explaining every change

Step 5: Summary Output

Present to user:

Original vs Optimized comparison (key changes)
Score improvements (Before → After for each dimension)
File locations

Key Principles

CAR Method for Bullet Points

Transform vague descriptions into CAR format:

Context: What was the situation?
Action: What did YOU specifically do?
Result: What was the measurable outcome?

Example transformation:

❌ "Responsible for managing team and improving processes"
✅ "Led 8-person logistics team (Context), implemented new WSSI forecasting system (Action), reducing stockouts by 35% and saving £120K annually (Result)"

ATS Optimization Rules

Use standard section headers: Experience, Education, Skills (not fancy variations)
Include full keywords from JD: If JD says "Supply Chain Optimization", use exact phrase
Avoid tables, headers/footers, graphics: ATS may not parse them
File format: .docx preferred over PDF for ATS

Quantification Guidelines

Always seek numbers:

Revenue: £X, $X, % growth
Scale: X team members, X regions, X SKUs
Efficiency: X% faster, X% cost reduction, X hours saved
Impact: X customers, X users, X% satisfaction improvement

Reference Materials

ATS Keywords: See references/ats_keywords.md for industry-specific keyword lists
Resume Templates: See references/resume_templates.md for UK professional format examples
Action Verbs: See references/action_verbs.md for strong starters

Output Format

The optimized resume should:

Maintain user's original structure (unless severely flawed)
Add quantifiable metrics where possible
Use CAR format for bullet points
Include all JD keywords naturally
Have Word comments on EVERY change explaining the rationale

Comment format in Word:

Location: [Section - Bullet Point]
Change: [Original → Modified]
Reason: [Why this improves the resume]
Evidence: [Based on user's answer / JD requirement / Best practice]

resume-analyzer