Built an AI-powered ATS Resume Builder using n8n
Over the last few days, I’ve been building an automation workflow that:
• Accepts a resume + job description
• Extracts resume content automatically
• Performs ATS analysis using AI
• Identifies missing keywords and weak areas
• Rewrites the resume dynamically based on the JD
• Generates a clean ATS-friendly PDF resume
• Optimizes formatting to stay within 2 pages
• Returns recruiter-optimized output automatically
Tech stack used:
- n8n forms
- Azure Open AI
- PDFBolt for HTML → PDF generation
- Prompt Engineering
One of the most interesting parts was solving:
- ATS keyword alignment
- PDF layout compression
- recruiter readability
- hallucination prevention
- dynamic JD adaptation
- compact HTML rendering
The workflow now behaves almost like an “AI Resume Architect” instead of just another ATS checker.
Current flow:
Upload Resume
→ Extract Text
→ ATS Analysis
→ Dynamic Keyword Extraction
→ AI Resume Rewrite
→ HTML Optimization
→ PDF Generation
→ Email Delivery
What surprised me most:
Prompt engineering mattered more than the AI model itself.
The biggest improvement came from:
- structured prompts
- layout optimization
- dynamic ATS target extraction
- recruiter readability constraints
This started as a simple automation experiment using n8n, but it’s slowly evolving into a real micro-SaaS idea.
Curious to know:
What features would YOU expect from an AI resume optimization platform?
