Why I Stopped Using Overleaf for My Resume (And Built a No-Code Alternative)

For software engineers and tech professionals, the LaTeX resume is almost a competency signal.

When a hiring manager scans a stack of resumes, a LaTeX document stands out instantly. The margins are mathematically balanced, the font (often Computer Modern) is crisp, and the bullet points align perfectly. More importantly, LaTeX compiles text natively, making it a dream for Applicant Tracking System (ATS) parsers that choke on Canva sidebars or Word text boxes.

But let's be honest about the developer experience: writing resumes in LaTeX is terrible.

Here is why I stopped using Overleaf for my resume, and how I built a no-code compiler to solve the problem.

The Pain of Raw LaTeX

Every time I wanted to update my resume, I went through the same cycle:

1. Logging into Overleaf.
2. Typos causing cryptic compiler errors like Missing } inserted or Paragraph ended before \multispan was complete.
3. Copy-pasting verbose blocks of code just to add a new side project.
4. Trying to squeeze my experience onto a single page by inserting hacky \vspace{-3pt} commands between lines.

I realized: I write code for a living, but I shouldn't have to compile my CV.

Building a No-Code LaTeX Engine

I wanted to combine the best parts of LaTeX (the typography and ATS-friendliness) with the ease of use of a visual form builder. The result is Lampzi (https://lampzi.com).

Here is the high-level architecture of how I built it:

1. The Form Builder: A React SPA that tracks resume sections (Education, Experience, Projects, Skills) as structured JSON objects.
2. The Express Backend: When the user clicks "Download," the React frontend posts the JSON data to our Node.js API.
3. Template Compilation: The backend injects the JSON values into clean, pre-designed .tex files.

4. pdflatex Execution: The server spawns a shell process to compile the document using a full texlive installation:
   bash pdflatex -interaction=batchmode -output-directory=tmp resume.tex

5. AI Importer: To make it even faster, I integrated the Gemini API. Users upload their existing PDF resume, and the AI parses the content, maps it to our JSON schema, and populates the form in seconds.

The Result

The output PDF is indistinguishable from a resume hand-coded in Overleaf. However, instead of taking hours, making a change takes 10 seconds on a phone or laptop.

If you're tired of wrestling with Overleaf packages just to apply for a job, check out Lampzi: https://lampzi.com

It is completely free to create and download. I'd love to hear your thoughts on the template aesthetics and compilation speeds!