Content Optimizer

Gradient descent for your career content.

What This Does

Build your career profile once. Generate resumes, cover letters, and other career content tailored to specific opportunities without copy-pasting or manual rewrites.

One source of truth (career.md) → many outputs.

Requirements

• Docker
• Python 3.13+

Setup

1. Build your career profile

Create career.md with detailed descriptions of your work, projects, education, and skills. This is your source of truth.

Go deep. If your algorithms course grade was a 67, write that. The richer your profile, the better the output, and the less likely you'll see hallucinated achievements that never happened.

2. Install

# Recommended - https://docs.astral.sh/uv/
uv sync

or

pip install -e .

Usage

With Claude Code (recommended):

1. Drop an opportunity file in opportunities/ - job description, role details, whatever
2. Ask Claude to generate content for it
3. Review the PDF in output/

Manual:

# Generate resume
uv run python -m cli render resume opportunities/example.yaml

# Generate cover letter
uv run python -m cli render cover-letter opportunities/example.yaml

# Recompile after editing the .tex file
uv run python -m cli re-render output/example.tex

(If using pip instead of uv: same commands, just drop the uv run prefix)

How It Works

You create opportunity-specific YAML files that select and frame content from your career profile. Templates render these to LaTeX, compile to PDF.

• Content changes? Edit the YAML, re-render.
• Formatting tweaks? Edit the generated .tex file, recompile.
• Want to customize templates? They're in templates/.

The CLI uses Pydantic schemas to validate structure. Check cli/schemas/ if you're curious.

---

If you find this useful, consider starring the repo.