# How to Vibe Code Your Own Rezi (and Stop Paying for It)

> The AI resume builder and career platform designed to beat applicant tracking systems.

- Site: https://rezi.ai
- Category: Career & Resume SaaS
- Verdict: **Solid side project** (75/100 vibecodeable)
- Estimated effort: 2-3 weeks part-time

## Verdict

Build a personal-use resume builder clone with AI bullet writing and keyword scoring in 2 to 3 weekends, but keep paying if you need their 4.5M user network or university white-labeling.

Rezi's core value—a structured resume editor combined with ATS-compliant PDF generation, a 23-metric scoring engine, and LLM-powered bullet point rewrites—is a textbook solo-developer SaaS project. While cloning their entire 10-year enterprise platform, job board aggregator, and Chrome extension ecosystem is a massive waste of time, building a private web app that ingests your work history, grades it against a job description using an LLM, and compiles a pristine single-column PDF is entirely achievable. The main engineering friction points will be fine-tuning the PDF export layout so it never breaks formatting rules and writing robust static-analysis heuristics for the resume scoring engine.

### What you can't replicate

- The 4.5 million registered user community and brand authority
- Enterprise university partnerships (CareerOS)
- Proprietary human expert resume review operations

## What it does

Rezi evaluates resumes across 23 key metrics, offers real-time content analysis for bullet points, matches job descriptions to inject target keywords automatically, and includes an AI resume agent, cover letter generator, and exportable machine-readable PDF pipelines.

### Core features

- Structured section-based resume editor (experience, education, projects, skills)
- Single-column ATS-compliant PDF generation engine
- Rezi Score multi-metric calculation engine (evaluating quantified metrics, filler words, passive voice)
- Job description keyword scanner and target extraction
- AI bullet-point writer and rewriter integrations
- Cover letter and resignation letter generators
- User authentication and database project isolation

## The business

### Pricing

- Free: Free — Basic resume creation and limited downloads.
- Pro: $29/mo — Full access to all templates and unlimited generations.

### Funding

$650,000 raised.
- Bootstrap & Regional Seed/Grant Funding
Investors: Startup accelerators, Government and institutional grants, Angel investors

Founded 2015.
Team size: 10-15.

## The hard parts

- Constructing a deterministic 23-point heuristic scoring engine that mirrors actual ATS parsing hurdles
- Implementing strict rich-text editing constraints (using Lexical or similar) that preserve layout integrity across exported PDFs
- Prompt orchestration and state management for real-time bullet-point content analysis without breaking UI responsiveness
- Managing PDF layout engines to guarantee zero multi-page spillover or formatting corruptions

## How to vibe code Rezi

### Prerequisites

- Node.js (free): Runtime environment for the Next.js full-stack application.
- GitHub (free): Source code control and deployment pipeline integration.
- Anthropic API Key (pay-as-you-go): Powers the AI resume writer, bullet-point rewriter, and keyword matching engine.

### Recommended AI tools

- Cursor: AI-native code editor ideal for rapid component authoring and iterative UI styling.
- Claude Code: Autonomous terminal agent for scaffolding complex database schemas, API routes, and PDF compilation pipelines.

### Stack

- Frontend: Next.js (App Router, Tailwind CSS, Lucide icons, React Hook Form)
- Backend: Next.js Server Actions & API Routes
- Database: Turso (SQLite at the edge via libSQL)
- Auth: better-auth
- Payments: None (Personal use clone)
- Other: Vercel AI SDK, pdfmake or react-pdf for server-side single-column resume generation

### Hosting

- Vercel (Hosting the Next.js frontend, server actions, and PDF compilation endpoints.): $0/mo (Hobby Tier)
- Turso (Serverless SQLite database storing user profiles, resume JSON structures, and section metadata.): $0/mo (Free tier)

### Build guide

1. **Project Scaffolding & Database Schema** — Initialize a Next.js project with Tailwind CSS, configure better-auth with SQLite via Turso, and establish database models for users, resumes, sections (experience, education, skills), and job target descriptions.

```
Create a new Next.js App Router project configured with Tailwind CSS, TypeScript, and Drizzle ORM connected to a Turso libSQL database. Set up better-auth for email/password authentication with database persistence. Implement the database schema for a resume-builder app: a `users` table, a `resumes` table (storing title, target job title, user_id, updated_at), and a `resume_sections` table (storing section type like experience/education/skills, order index, and structured JSON content payload). Ensure all relations and foreign keys are strictly enforced and migration scripts are generated and tested locally.
```

2. **Structured Resume Editor Interface** — Build a responsive split-screen editor UI with a structured form on the left for entering professional summary, work experience items, bullet points, and skills, alongside a live preview canvas on the right.

```
Build a responsive split-screen resume editor page in Next.js. The left side must provide structured form controls for managing resume sections: contact information, professional summary, work experience (company, title, date range, bullet points list), education, and skills tags. The right side must feature a clean, high-density, single-column document preview canvas styled to mirror standard ATS-safe resume formatting. Connect the form inputs to local React state with automatic debounced autosave calling a Next.js server action to persist changes into the resume_sections JSON payload table.
```

3. **ATS-Compatible PDF Generation Engine** — Implement a server-side PDF generation pipeline using `pdfmake` or `react-pdf` that compiles the structured resume JSON into a clean, single-column machine-readable PDF document.

```
Implement a server-side PDF generation endpoint using `pdfmake` (or `react-pdf`) that compiles a user's resume JSON record into a downloadable, text-selectable, single-column PDF document. Ensure strict adherence to ATS compatibility standards: standard standard font families (Helvetica/Times), single-column tabular layouts, zero hidden graphics or multi-column text frames, clear section headings, and proper margins. Expose a download button in the editor toolbar that triggers this server action and streams the PDF file directly to the client browser.
```

4. **Rezi Score Multi-Metric Evaluation Engine** — Create a deterministic analysis engine that evaluates resume content against heuristic rules (checking for quantifiable metrics, passive voice, bullet point punctuation, and filler words) to output a composite score and section breakdowns.

```
Create a deterministic resume analysis service in TypeScript that evaluates a resume JSON payload against 23 scoring heuristics split into categories: Content (checking for quantified numbers, action verbs, filler words, passive voice), Format (checking contact info completeness, section headers), and Best Practices. Implement rule functions that scan bullet strings and return itemized suggestions (e.g., 'Add a quantified metric to this bullet point'). Build a React component dashboard widget displaying the overall 'Rezi Score' out of 100 with category badge breakdowns and actionable improvement suggestions.
```

5. **AI Keyword Targeting & Bullet Point Writer** — Integrate the Vercel AI SDK and Anthropic API to allow users to paste a job description, extract missing target keywords, and automatically rewrite or generate impactful bullet points.

```
Integrate the Vercel AI SDK and Anthropic API to power two core AI features: 1) Job Keyword Targeting: a component where users paste a job description, and the LLM extracts key technical/soft skills, comparing them against the current resume content to highlight matched vs missing keywords with a '1-click add' action. 2) AI Bullet Writer: an inline button on any work experience entry that takes the user's rough draft note and rewrites it into an action-oriented, results-driven bullet point adhering to resume best practices. Use `generateObject` with strict Zod schemas for reliable JSON responses from the model.
```

6. **Cover Letter Generator & Final Polish** — Add a companion cover letter generator tool that pulls data from the user's active resume and target job description, and perform end-to-end testing and bug fixing.

```
Build a cover letter generator page that accepts a target job description and links to an existing resume record. Use an Anthropic API call via server action to synthesize the candidate's professional history into a tailored, professional cover letter formatted cleanly for export or clipboard copying. Add error boundaries, toast notifications for autosave and export actions, and refine the Tailwind UI to match a polished SaaS aesthetic. Test the complete flow from authentication, resume editing, AI keyword targeting, scoring, down to PDF export.
```

### Cost vs paying

**Starting costs (one-time):**

- Custom domain (optional): $12/year
- Anthropic API developer credits: $10 one-time
- Total: ~$22 one-time

**Ongoing costs (monthly):**

- Vercel Hobby & Turso Free Tier: $0/mo
- Anthropic API usage for AI bullet writing and keyword matching: ~$3-5/mo
- Total: ~$4/mo

- Paying for the SaaS instead: $29/mo (Pro Plan)
- Build time: 25-35 hours
- AI tool credits: $20/mo (Cursor Pro / Claude Code)
- Break-even: Immediate (if built for personal use and replacing the $29/mo subscription)

## Sources

- [Rezi Official Website & Feature Pages](https://rezi.ai)
- [Rezi Pricing & Plan Comparison Matrix](https://rezi.ai/pricing)
- [Rezi About & Company History](https://rezi.ai/about)