YAMLResume VS Reactive Resume

Reactive Resume is the most popular open-source resume builder on GitHub, created by Amruth Pillai. It has been around since 2020 and has grown to 43,000+ GitHub stars, with a hosted version at rxresu.me that anyone can use for free. The codebase is open source under the MIT license, with no tracking, no ads, and no paywalls.

From its official repository:

Reactive Resume is a free and open-source resume builder that makes it easy to
create, update, and share your resume.

- Pick a template, fill in your details, and export to PDF.
- Basic use needs no account.
- You own your data. The codebase is open source under the MIT license, with
  no tracking, no ads, and no hidden costs.

Reactive Resume began as a side project and was first released in early 2020. It reached v2 later that year, v3 by early 2022, and v4 in late 2023 with a NestJS and PostgreSQL backend. In January 2026, v5 was released—a complete rebuild on TanStack Start (React 19, Vite, TypeScript, PostgreSQL with Drizzle ORM) with an AI agent, an MCP server, and a job application tracker. Later that year, the stack shifted to TanStack Router with Hono and oRPC, and the server-side printer service was replaced by client-side PDF rendering with @react-pdf/renderer.

Reactive Resume and YAMLResume share the same values—free, open-source, privacy-respecting resume tooling—but they embody opposite workflows. Reactive Resume is a web application: you edit your resume in a visual, WYSIWYG builder and your data lives in the app (PostgreSQL, or your self-hosted instance). YAMLResume is a file-based toolchain: your resume is a plain-text YAML file that you edit in any editor, version-control with Git, and compile to multiple output formats from the command line. Here is how they compare:

Feature Comparison

Key Differences Explained

If we have to choose one phrase to summarize the difference between YAMLResume and Reactive Resume, it would be files versus apps. YAMLResume treats your resume as plain-text source code; Reactive Resume treats it as data inside an application. Nearly every other difference flows from that decision.

Files vs. Apps

With YAMLResume, a resume is a YAML file on your disk:

$ yamlresume build resume.yml
✔ Generated resume pdf file successfully: resume.pdf
✔ Generated resume docx file successfully: resume.docx
✔ Generated resume markdown file successfully: resume.md
✔ Generated resume html file successfully: resume.html

You can commit it to Git, diff two versions, branch per job application, review changes in pull requests, and let CI compile the artifacts. The GitHub Action can build and attach PDFs to releases automatically. Nothing runs in the background, and nothing leaves your machine except the files you choose to publish.

With Reactive Resume, a resume lives inside the application. You edit through the web UI, the data is stored in PostgreSQL (on the hosted service or your self-hosted instance), and you export artifacts when you need them. The app offers JSON export/import, snapshot-based version history with restore (captured at milestones such as imports and AI edits), and full data deletion—but there are no line-level diffs or branches; if you want those, you must build that workflow yourself on top of JSON exports.

Both approaches are legitimate. The app model removes all setup and is the fastest path from zero to a finished PDF; the file model pays off over time as your resume accumulates history, variants, and automation.

Data Model

Unlike RenderCV, which allows arbitrary section titles, both YAMLResume and Reactive Resume use a fixed, structured schema with well-known sections. That makes resumes from either tool predictable and machine-readable.

YAMLResume's content model separates three concerns:

• content: the resume itself (basics, location, profiles, work, education, skills, projects, publications, awards, certificates, volunteer, references, languages, interests)
• locale: the language and region the resume is written for
• layouts: presentation settings (template choice, section aliases and ordering, typography, page setup)

Because the data model is fixed and aligned with JSON Resume, YAMLResume can validate it rigorously, normalize values into fixed vocabularies (education degrees, skill levels, language fluency, countries and regions translated into every supported language), and interchange data with the broader JSON Resume ecosystem. The compiler even accepts JSON Resume files directly, and the json2yamlresume converter handles full projects.

Reactive Resume's schema covers similar ground (basics, experience, education, skills, projects, certifications, languages, interests, references, awards, volunteer work, publications, profiles) and adds capabilities YAMLResume does not have:

• Custom sections: define entirely new sections with typed extension fields, useful for portfolios, talks, or patents
• Per-item visibility: hide individual entries without deleting them, so one resume can produce many tailored variants
• Pages as artboards: drag entries between pages and see exactly where page breaks land
• Cover letters: standalone letters styled from a resume, exported to PDF, DOCX, or Markdown
• Application tracker: a job-search pipeline with stages, notes, follow-ups, and attached documents, also reachable through the REST API and MCP tools

Reactive Resume also imports JSON Resume files (with validation that rejects impossible dates), so moving between the two ecosystems is possible from both sides.

Editing Experience

Reactive Resume provides a polished WYSIWYG editor: what you see in the browser is what lands in the PDF. Every text field supports a rich text editor with bold, italic, links, lists, and even highlight colors. For most people this is the friendliest possible way to write a resume—no syntax to learn, no build step to run.

YAMLResume deliberately trades some of that immediacy for portability and scale. You write YAML in your own editor, and a constrained set of Markdown syntax (bold, italic, links, lists, paragraphs) is supported in summary fields. The payoff:

• Your resume is grep-able, scriptable, and diffable
• Multi-line YAML strings keep long summaries readable
• The same content compiles to LaTeX, HTML, DOCX, and Markdown
• Bulk edits (renaming an employer across ten variants) are a sed away

Layouts and Rendering

The rendering architectures diverge completely.

Engines

YAMLResume provides four rendering engines:

• DOCX engine: Microsoft Word documents for easy editing and sharing
• HTML engine: Responsive, web-ready resumes with SEO metadata
• Markdown engine: Clean text for LLMs, ATS systems, and static sites
• LaTeX engine: Pixel-perfect PDFs with professional typography via XeTeX

A single yamlresume build command generates all four formats from the same source file.

Reactive Resume renders PDFs in the browser with @react-pdf/renderer (v5 launched with a server-side printer service; it was removed in v5.1). Exports include PDF, DOCX, Markdown, and JSON. Because the builder preview is rendered with pdfjs against the same document tree used for export, output is deterministic and page breaks preview exactly as artboards. The trade-off: PDF and DOCX are the main document outputs, with Markdown export added in v5—there is no LaTeX pipeline and no standalone HTML export.

Design customization

Reactive Resume offers the broader visual toolkit:

• 15 templates (Azurill, Bronzor, Chikorita, Ditgar, Ditto, Gengar, Glalie, Kakuna, Lapras, Leafish, Meowth, Onyx, Pikachu, Rhyhorn, and Scizor), mostly named after Pokémon
• Per-template options: colors, fonts, sizes, spacing, A4 or Letter, sidebar width, full-width pages, icon styles for skill/language indicators
• Semantic CSS: a CSS-like stylesheet language whose selectors describe resume content (section[type="experience"], field[name="position"], list-marker) rather than template markup, so styles survive template switches and copy between resumes
• Drag-and-drop arrangement of sections, items, and pages

YAMLResume's LaTeX engine provides page customization (paper size, margins, page numbers) and typography controls (font family, font size, line spacing), while templates encapsulate the visual decisions—per-element color and fine-grained spacing control are intentionally absent. If you want to obsess over every pixel of a unique design, Reactive Resume gives you more levers. If you want a professionally designed result without becoming a typesetter, YAMLResume's constrained approach is the point.

Section customization

YAMLResume reshapes structure through section aliases and section reordering without touching the data—rename work to "Employment History" or move skills above education in a layout config.

Reactive Resume achieves similar results interactively: sections can be renamed, reordered with drag-and-drop, and extended with custom sections. Because it is app data rather than source code, these changes are immediate but live outside version control unless you export JSON.

Internationalization

The two tools localize at different layers.

Reactive Resume localizes the interface: the app UI is translated into 55 locales via Crowdin, and the PDF pipeline handles right-to-left mirroring across all templates with locale-aware fallback fonts for Korean, Japanese, Chinese, Arabic, Hebrew, and Thai. However, the resume content carries no localization layer: section headings come from the template, dates follow one format, and vocabulary like degrees or skill levels is whatever you type. A French resume means writing French content by hand.

YAMLResume localizes the content. Setting the locale key switches section headings, country names, education degrees, skill levels, language fluency levels, idiomatic date and address formats, locale-appropriate punctuation, and LaTeX babel configuration across 10 languages and 12 locale codes. A single source file can be compiled per language, and yamlresume ai translate produces structure-preserving translations—so maintaining an English and a Chinese resume is one file plus a translation step, not two parallel documents drifting apart.

For CJK and complex scripts, YAMLResume's LaTeX engine also brings decades of mature typesetting support—ctex for Chinese, luatexja for Japanese, kotex for Korean, and xeCJK for XeLaTeX—handling font fallback, full-width punctuation compression, CJK-Latin spacing, and locale-aware line breaking.

AI Workflows

Both projects embraced AI, in characteristically different ways.

Reactive Resume embeds AI in the app: bring your own API key from providers including OpenAI, Anthropic, Google Gemini, OpenRouter, and Ollama (plus any OpenAI-compatible endpoint) and an in-app agent can revise content, fix grammar, and adjust tone, with an action history you can restore. Its ATS check is deliberately deterministic—evidence-based feedback with an optional AI second opinion that returns no score of its own. AI also powers resume import from PDF or DOCX, reconstructing structured data from existing documents. Through the MCP server, agents can manage resumes, cover letters, and the application tracker end to end.

YAMLResume embeds AI in the pipeline:

• yamlresume ai generate: bootstrap a complete, schema-validated resume from a job title and locale
• yamlresume ai translate: translate a resume while preserving its structure and layout configuration

Both run from the terminal or CI, store nothing server-side, and feed results into the same validated file workflow. Combined with the Markdown engine, a YAMLResume file is also one of the most LLM-friendly resume formats around—readable context for any coding agent, without a dedicated integration.

APIs and Automation

Reactive Resume exposes a remote API surface: an OpenAPI-documented REST API, an MCP server with OAuth, passkeys, and two-factor authentication. It is a server integration—you authenticate against an instance.

YAMLResume exposes a local API: @yamlresume/node reads, validates, builds, watches, generates, and translates files from any Node.js application, plus create-yamlresume to scaffold projects. Combined with the GitHub Action, Docker image, and plain files, it automates without standing up a service.

Ecosystem and Community

Self-hosting vs. Local-first

Reactive Resume is self-hostable: an all-in-one Docker image (Docker Hub amruthpillai/reactive-resume, GHCR ghcr.io/reactive-resume/reactive-resume) and a one-click Vercel deployment that provisions Neon PostgreSQL, Upstash Redis, and Vercel Blob storage. Self-hosting gives you full control, but you are operating an application: database migrations, storage, authentication, and backups are yours to run.

YAMLResume is local-first: everything runs on your machine, and the official Docker image exists only to provide a preconfigured build environment (Node.js, XeTeX, fonts). There is no server to operate, no database to back up, and no account to secure.

Distribution

YAMLResume meets developers where they work:

• npm/pnpm/yarn/bun: npm install -g yamlresume
• Homebrew: brew install yamlresume on macOS and Linux
• GitHub Action: build resumes in CI/CD
• Docker: one-command builds anywhere

Reactive Resume distributes as the hosted app plus container images and source; there is no CLI, Homebrew formula, or CI integration.

JSON Resume Interop

Both tools can import JSON Resume files, which lowers the cost of switching in either direction. YAMLResume goes further: its compiler accepts JSON Resume natively, the schemas stay aligned, and json2yamlresume converts whole projects including layouts.

Web Interface

YAMLResume provides an interactive, web-based playground that runs entirely in the browser with no server-side rendering—useful for trying the format without installing anything.

Reactive Resume's "playground" is the product itself: a polished hosted app at rxresu.me, free without an account for basic use.

Summary

Reactive Resume excels if you want:

• A visual, WYSIWYG resume builder with zero setup and no build step
• 15 polished templates with deep visual customization, including a Semantic CSS editor and drag-and-drop layout control
• WYSIWYG rich text editing across every field
• Public share links with view counts
• Built-in cover letters and a job application tracker with pipeline stages
• An in-app AI agent supporting OpenAI, Anthropic, Google, OpenRouter, and Ollama, AI import from PDF/DOCX, and an ATS check
• Agent and automation integrations through a hosted MCP server and OpenAPI REST API
• A self-hosted web application you can deploy with one click on Vercel or via Docker

YAMLResume excels if you want:

• Your resume as a plain-text YAML file in Git, with native diffing, branching, and code review
• Four rendering engines (DOCX, HTML, Markdown, LaTeX/PDF) from a single source file
• Deep content internationalization: 10 languages across 12 locale codes, with translated vocabularies, idiomatic dates and addresses, and CJK-ready LaTeX typography
• A CLI-first developer experience with watch mode, schema-aware diagnostics, and deterministic builds
• AI generation and translation that run locally in your terminal or CI
• Ecosystem tooling built for automation: Homebrew, Docker, GitHub Action, a Node.js API, and an embeddable browser playground
• No accounts, no servers, no databases—just files

Choose Reactive Resume if you prefer a polished visual app that manages your resumes, cover letters, and job search in one place, and you want the richest template and AI-agent experience without touching a command line.

Choose YAMLResume if you think of your resume as source code—versioned, reviewable, compilable, and automated—and you want one file to produce a pixel-perfect PDF, a Word document, a web page, and an LLM-friendly text version, in a dozen locales, on every commit.