Learning and careers decision
Rezi
A single developer can reproduce the core resume-writing and export workflow (LLM prompts, keyword matching, PDF/DOCX exports) in about a week, but Rezi claims proprietary resume-trained models and enterprise features (SSO, webhooks, paid reviews) that are not practical to fully replicate.
Visit website↗Open-source builds that already do this
Every project below is open source and already does this job today. Fork one, self-host it, or take the parts you need — the build prompt further down assumes an empty file, and this is the shortcut past that. Licences differ; check the one on each card before you ship.
What a replacement has to do
- Allow a user to upload or enter resume content, run an AI prompt to generate/ rewrite bullet points and summary targeted to a pasted JD, compute an ATS-style score & missing keywords, and export the result to PDF/DOCX.
What it still won’t have
- Rezi-trained resume models and any proprietary fine-tuning
- Enterprise features and operational integrations (SSO, webhooks, team management)
- Large user-scale infrastructure, analytics and trust signals from millions of users
- Human-paid resume review service
What remains hard
- Proprietary models
Unlike generic AI models like ChatGPT , Rezi’s AI models are specifically trained for writing resume content.
First-year cost
Keep paying
Paying is—cheaper in year one.
On cash alone, building overtakes the subscription at 3 seats.
Money you would actually spend
Time you would spend
—
What you would spend
What we assumed
The verdict above measures whether you could build it. This one is only about money.
Runnable build prompt
Build a minimal AI resume web app using Node.js + Express, React for the frontend, Postgres for storage, and OpenAI-compatible LLM API calls. In scope: (1) an upload or LinkedIn import page that parses resumes into structured experience/skills entries; (2) an endpoint that accepts a job description and computes keyword matches and a simple match score; (3) an endpoint that calls an LLM to generate/ rewrite bullet points and a 2–3 line summary targeted to the job description; (4) ATS-safe HTML templates and export to PDF and DOCX; (5) basic user accounts and one-seat subscription flag. Out of scope: enterprise SSO, paid human resume review, analytics dashboards, and large-scale multi-tenant orchestration. Include input validation, error handling, unit tests for parsing and scoring logic, and end-to-end tests for the upload → generate → export flow.
How we checked
How the score was reached
- Partly verdict base52
- An open-source build was found+5
- 5 cited sources+3
- Price verified on pricing page+3
- 3/3 assessment runs agreed+4
- Hard moats found in the evidence-3
- Evidence score64
The base comes from the verdict. Everything under it is a check that either happened or did not, and each one is a fact frozen in this record rather than a judgement made at render time — so the same evidence always produces the same number.
How scoring works →Cited sources · 5
Every page the run actually retrieved.
- official productRezi product page
- official pricingRezi pricing
- official docsRezi docs / user guides
- open sourceLingyiChen-AI/JadeAI
- open sourcesrbhr/Resume-Matcher
Integrity checks
What held up, and what did not.
