AI-Assisted Resume Builder helps you create tailored, ATS-friendly resumes for every job application with minimal effort. Instead of manually rewriting your resume for each position, this tool uses AI to automatically match your experience with job requirements and company values.
-
👤 Build Your Profile Once: Create a comprehensive
profile.jsonfrom your existing resume, conversation, or structured input. This becomes your single source of truth for all professional experience. -
🔍 Match Any Job: Paste a job description, and the AI analyzes keywords, requirements, and company values. It scores how well your experience matches the position and identifies skill gaps.
-
⚡ Generate Tailored Resumes: The AI automatically:
- Selects your most relevant experiences (top 3-5 based on match scores)
- Rewrites bullet points using the STAR method (Situation, Task, Action, Result)
- Emphasizes job-relevant keywords and technologies naturally
- Incorporates company values and culture signals from the job description
- Preserves quantifiable metrics and achievements
- Formats everything into professional LaTeX for ATS compatibility
-
📄 Export to PDF: Compile your tailored resume into a polished PDF ready for submission.
- Truthful & ATS-Optimized: All content is grounded in your real experience (
profile.json), ensuring authenticity while maximizing keyword alignment for applicant tracking systems. - Position-Specific: Each resume variant is optimized for a specific role, highlighting the most relevant achievements and skills.
- Company-Aware: The AI considers company values and culture mentioned in job descriptions, helping you align your resume with what the organization values.
- Efficient: Generate multiple tailored resumes in minutes instead of hours of manual work.
For a quick start with pre-configured settings:
-
Download for your editor
- Go to Releases
- Download the package for your editor (Cursor, VS Code, JetBrains, etc.)
- Extract and open in your AI-assisted IDE
-
Install a LaTeX distribution (required for PDF generation, choice depends on template)
- Recommended full distributions (include most packages out of the box):
- Default template dependencies:
moderncvclass plus common packages likegeometry,multicol,ragged2e,hyperref,inputenc,fontenc
For full control to modify commands and customize to your requirements:
-
Clone the repository
git clone https://github.com/yourusername/ai-resume-builder-agent.git cd ai-resume-builder-agent -
Customize commands (optional)
- Edit command templates in
templates/commands/to match your specific needs - Modify prompts, workflows, or add custom commands
- Edit command templates in
-
Install a LaTeX distribution (required for PDF generation, choice depends on template)
- Recommended full distributions (include most packages out of the box):
- Default template dependencies:
moderncvclass plus common packages likegeometry,multicol,ragged2e,hyperref,inputenc,fontenc
-
Open in your AI Editor Open the folder in Cursor, VS Code (with Copilot), or your preferred AI-assisted IDE.
profile.json: Your source of truth for professional experience (generated bybuild-profile).resumes/: Directory where tailored resumes are generated, organized by company/role.resume_templates/: LaTeX templates used for styling resumes.templates/commands/: System prompts that power the AI agent commands.
Follow these 4 simple steps:
Use the build-profile command to build or update your professional profile from resume text, conversation, or structured input.
Use the match-job command to analyze a job description and identify the most relevant parts of your experience.
Use the generate-resume command to generate a tailored resume based on the latest match-job session.
Use the export-to-pdf command to compile your resume into a professional PDF ready to send.
Build or update your professional profile from resume text, conversation, or structured input.
Key features:
- Extracts contact info, work experience, education, skills, and projects
- Validates and formats into structured JSON
- Saves to
profile.jsonin repository root
Use when: First-time setup, profile updates, or consolidating multiple sources
Analyze a job description to identify keywords and score how well your experience matches.
Key features:
- Accepts URL, pasted text, or file path
- Extracts keywords, skills, and requirements
- Scores experience relevance (exact matches: 3pts, related: 2pts, soft skills: 1pt)
- Creates match-job session in
generated/<sequence>-<company>-<position>/ - Shows relevance report with ranked experiences and skill matches
Use when: Before applying to understand alignment and identify skill gaps
Generate a tailored LaTeX resume highlighting your most relevant experience.
Key features:
- Uses latest match-job session and
profile.jsonprofile - Selects top 3-5 relevant experiences, rewrites summary with JD keywords
- Transforms bullets to STAR format (Situation, Task, Action, Result)
- Preserves quantifiable metrics and emphasizes job-relevant keywords
- Generates
<position>_resume.texinresumes/<company>/<position>/
Use when: After running match-job to create position-specific resume variants
Compile the generated LaTeX resume into a professional PDF.
Key features:
- Checks environment and verifies LaTeX installation
- Prompts to install missing packages if detected
- Locates latest resume from
resumes/<company>/<position>/ - Compiles with
pdflatex(runs twice for references) - Generates
<position>_resume.pdf - Optionally cleans auxiliary files
Use when: After generating a resume to get final PDF for submission
Import a custom LaTeX resume template from text or URL.
Key features:
- Accepts URL, pasted code, or file
- Validates LaTeX structure
- Saves to
resume_templates/directory
Use when: You want custom styling or multiple template options
Contributions are welcome! Please feel free to submit a Pull Request.
This project is inspired by and uses concepts from spec-kit, a toolkit for Spec-Driven Development created by GitHub. The command template structure and workflow patterns follow similar principles of structured, AI-assisted development workflows.
This project is licensed under the MIT License - see the LICENSE file for details.