ten-seconds-cv

Hi! My name is Esteban Martinena, a Software Engineer from Spain. This is my resume, built with LaTeX, Pandoc, Docker...

Download

Two formats are available. Plain first, then two-columns:

• Plain — single-column, serif-based, reference-inspired design.
• Two-columns — sidebar layout with skill heatmap, based on AltaCV.

Both share the same data source (data.md). Format-specific LaTeX templates live under latex/plain/ and latex/two-columns/.

Why?

As many other developers, I really hate to write or update my resume. I've tried many different approaches, but I've never been satisfied with the result. Including new information, changing the format, or even modifying the order of the sections was always a pain.

At some point, I decided to write my resume in LaTeX, but the more important thing was to automate the process of building the PDF file and extracting the information from a single source of truth.

Also, I wanted to have a repository with my resume, so I could track the changes and share it with others, and at the same time show some of my skills as DevOps and Software Engineer here.

So, not only the PDF generated in this repository is my resume, but also the repository itself.

How?

Apart from using some nice LaTeX templates, I've used Pandoc to convert the Markdown file to LaTeX. This way I can write my resume in the data.md file, and then generate the LaTeX file and the PDF when I change it.

I've also used Docker to build the PDF file, so I (or anyone else) can build it without installing anything else.

Data fields are shared across formats. Optional fields like summary, expertise_groups, and featured_experiences are used by the plain template, with fallbacks to the legacy fields for two-columns.

I've also created a simple Python script (heatmap.py) that generates an EPS heatmap of my skills, used by the two-columns format.

GitHub Actions

Docker image build

To avoid building the Docker image every time, a workflow builds it and pushes to GHCR. It's only triggered when the Dockerfile or requirements.txt changes.

PDF build

Runs on every push. It pulls the Docker image from GHCR and runs make pdf-all inside it, which builds both formats. The resulting PDFs are uploaded as build artifacts and, on tag/release, as release assets. The download links above point to the latest release assets.

How to build locally?

Requires Docker. Clone the repo and run:

./build.sh         # builds both formats
./build.sh plain   # plain only
./build.sh two-columns  # two-columns only

Outputs cv-esteban-martinena-{format}.pdf and cv-esteban-martinena-{format}-cover.png in the current directory.

References

The initial idea for this repo came from the following repositories:

1. Ten Seconds CV
2. Carmine Spagnuolo's Twenty Seconds Curriculum Vitae