The Problem
Recruiters expect a customized CV for each application, but manually editing LaTeX sources, re‑formatting, and tracking changes quickly becomes error‑prone and time‑consuming. Without a repeatable workflow, candidates cannot reliably align bullet points with job‑specific keywords or maintain version history.
What This Does
The repository treats a CV as code. The source LaTeX file v1/master.tex is the single source of truth for content and layout. Prompt files such as prompts/jobdescmatch.md provide ready‑made instructions for AI agents (Claude, GPT‑4, etc.) to rewrite or augment bullet points based on a target posting (v1/canva/jobdesc.md). The Docker‑based dev container defined in .devcontainer/devcontainer.json installs a full TeX Live environment, so contributors can compile the CV without local LaTeX setup.
Key artifacts:
v1/master.tex – base CV template (content + styling). v1/canva/main.tex – an example of a job‑specific variant generated after applying the agent prompt. prompts/jobdescmatch.md – the exact prompt text that drives the “match CV to job description” workflow.
The repo does not include any custom build scripts; compilation relies on standard LaTeX tools available inside the container.
How To Use It
Environment – Install Docker Desktop and the VS Code Remote‑Containers extension. Open the container – Clone the repo, open the folder in VS Code, and click Reopen in Container. VS Code will apply .devcontainer/devcontainer.json, which provisions TeX Live. Edit the master CV – Modify v1/master.tex as needed. The LaTeX Workshop extension (or any LaTeX tool inside the container) can compile the file:
cd v1 pdflatex master.tex # produces master.pdf Generate a job‑specific version – Place the target posting in v1/canva/jobdesc.md. Feed prompts/jobdescmatch.md together with master.tex into your preferred coding agent. Paste the agent’s output into v1/canva/main.tex. Compile main.tex to verify the PDF (pdflatex main.tex).
No additional configuration files, environment variables, or secret keys are required.
Real‑World Use
A candidate maintains master.tex as their canonical resume. For a data‑science role, they copy the posting into v1/canva/jobdesc.md, run the supplied prompt through Claude, and obtain a revised bullet list that emphasizes “machine‑learning pipelines” and “SQL performance tuning”. After pasting the result into v1/canva/main.tex and compiling, they have a PDF ready to upload, with the entire change tracked in Git.
Code Health & Issues
Medium – No test suite – repository lacks any test. files or testing framework configuration. Medium – No CI/CD – no .github/workflows or other pipeline definitions; builds are manual. Medium – Missing LICENSE – legal reuse and redistribution are unclear. Low – Limited documentation – only a README and a single prompt file; no usage script or Makefile to standardize compilation. Low – Potential platform lock‑in – the workflow assumes access to a commercial coding‑agent API; no fallback instructions are provided.
The Bottom Line
The repo offers a minimal but functional skeleton for treating a LaTeX CV as version‑controlled code, with a concrete AI‑prompt for keyword matching. It is suitable for technically proficient individuals who can operate a Docker dev container and run LaTeX manually. However, the lack of automated tests, CI, and licensing means the project is best used as a personal prototype rather than a production‑grade tool for teams.