Tailor Resume

Tailor Resume is a CLI tool that takes a base resume template and a job posting, then produces a tailored resume and a structured report explaining what changed.

The system is designed for iterative learning: you can refine scoring, structure, and content over time without rewriting the pipeline.

Mental Model

Think of Tailor Resume as a three-stage pipeline:

1. Capture
2. Transform
3. Export

1. Capture

---

Inputs:

• A base resume template (Markdown)
• A job posting (clipboard or file)
• Configuration (TOML + YAML)

Responsibilities:

• Capture job URL and description
• Parse job metadata (LinkedIn or generic)
• Enforce required fields (job title, company)
• Load configuration and stopwords

Important principle: The base resume is a TEMPLATE, not your real resume. It may contain placeholders such as:

YOUR_NAME_HERE {{CONTACT_LINE}}

Those placeholders must be resolved before export.

2. Transform

---

This is the core logic of the system.

Responsibilities:

• Parse the resume into structured sections
• Score bullets against the job description
• Retain, drop, or promote bullets per role
• Apply guardrails (e.g., migrations must appear if job mentions them)
• Reorder CORE COMPETENCIES based on relevance
• Inject contact information
• Add resume-specific YAML frontmatter

Key principles:

• JobPost.title remains the canonical job title in code
• YAML frontmatter uses job_title for clarity
• All transformations are deterministic and reproducible
• Config controls behavior; CLI flags override only when necessary

3. Export

---

Outputs:

• Job post Markdown (archival)
• Tailored resume Markdown
• JSON report explaining scoring and decisions
• Optional CSV run log
• Optional DOCX export via Pandoc

Important principle: Export scripts (Pandoc, DOCX) do NOT mutate content. They assume the Markdown is already final.

Directory Structure

Outputs are organized automatically:

out/ 01-January/ 09-Acme Corp/ YYYY-MM-DD_jobpost_acme_corp.md YYYY-MM-DD_resume_acme_corp_service_manager.md YYYY-MM-DD_report_acme_corp_service_manager.json

The output root and CSV log path are defined in config, not hardcoded.

Configuration

Configuration lives in tailor_resume.toml and supporting YAML files.

Examples:

• Contact information (mandatory)
• Output root directory
• CSV log location
• Scoring weights
• Guardrails
• Stopwords (global + per-company)

Stopwords are split intentionally:

• Global stopwords live in stopwords.yaml
• Company-specific stopwords are added per run

What This Tool Is (and Is Not)

This tool IS:

• A resume tailoring engine
• A learning platform for resume optimization
• A deterministic system you can refactor safely

This tool is NOT:

• A GUI application
• A one-click resume generator
• A final product (it is meant to evolve)

Typical Usage

1. Edit the base resume template
2. Run the CLI to capture a job posting
3. Review the tailored Markdown
4. Review the JSON report
5. Export to DOCX if needed

Philosophy

This project prioritizes:

• Clarity over cleverness
• Explicitness over magic
• Config over hardcoding
• Reproducibility over convenience

Refactors are expected. Learning is expected. V1 is intentionally conservative.