The Problem We Set Out to Solve
Job seekers waste hours every week manually tailoring their CVs for each role. Copy-paste, rewrite, repeat. And despite all that effort, the hard truth is that more than 75% of CVs are filtered out by Applicant Tracking Systems (ATS) before a single human ever reads them. The problem isn't lack of effort — it's the wrong tools.
We'd experienced this firsthand. Watching friends and colleagues send dozens of applications into a void, getting no responses, wondering what they were doing wrong. The CV they spent hours polishing was being silently rejected by software that couldn't even read a two-column layout correctly.
That's the problem CVwise was built to fix.
Starting Simple, Growing Fast
The first version of CVwise was a focused tool: upload your CV, paste a job description, get an optimized version back. Simple, fast, genuinely useful. Early users responded well — the feedback was consistent: "I finally got a callback after using this."
But as we talked to more users, the requests piled up. "Can I build my CV from scratch?" "Can you write my cover letter too?" "I want to save different versions for different industries." Over several months, CVwise grew from a single optimizer into a complete job application suite.
Built on Expert Insights and Real Data
CVwise wasn't built on guesswork. Before writing a single line of AI prompt, we consulted with senior HR professionals, hiring managers, and career coaches across multiple industries — finance, technology, consulting, and healthcare. Their input shaped every decision: what ATS systems actually penalize, what cover letter structures generate callbacks, what formatting kills an otherwise strong application.
We also analyzed over 1,000 top-tier CVs — applications that resulted in interviews at Fortune 500 companies, top consulting firms, and investment banks. We studied what patterns they shared, what language they used, how they structured experience bullets, and how they balanced brevity with depth. That analysis is embedded directly into CVwise's AI prompts and optimization logic.
The result is a tool that doesn't just rephrase your CV — it applies the same principles that demonstrably work at the highest level of the job market.
The Technology Stack
CVwise is built on Next.js 16 with the App Router, Supabase for authentication and data storage, Stripe for subscriptions, and Anthropic's Claude AI at the core of every intelligent feature. Here's how the AI architecture works:
- Claude Haiku handles fast parsing tasks — extracting structured data from uploaded PDFs and DOCX files, including multi-column layouts and scanned documents using Claude's vision capabilities.
- Claude Sonnet powers the high-quality optimization passes — rewriting CV sections to match job description keywords naturally, generating cover letters, and producing ATS keyword gap analysis.
This two-model approach keeps costs down while delivering genuinely impressive output quality where it matters most.
The Four Core Features
CV Builder — Guided 5-Step Form
Not everyone has an existing CV, or they want to start fresh. The CV Builder walks users through a structured five-step form: personal information and contact details, work experience with rich bullet-point editing, education history, skills and languages with proficiency levels, and a live preview that updates in real time. The result is a clean, well-structured CV ready for AI optimization.
Upload Existing CV
For users with an existing CV, they can upload a PDF or DOCX file directly. Claude's vision-enhanced parsing handles everything from standard single-column CVs to complex multi-column agency formats. Even scanned PDFs are processed with high accuracy.
AI Optimizer with ATS Scoring
This is the heart of CVwise. Users paste a job description URL (from LinkedIn, Indeed, Glassdoor) or paste the text directly. Claude Sonnet reads both the CV and job description, identifies keyword gaps, and rewrites sections to include exact keywords naturally — not stuffed awkwardly. The result includes a 0–100 ATS keyword match score so users can see exactly how well their CV aligns with the role before applying.
Cover Letter Pro
Research consistently shows that cover letters matter. CVwise's Cover Letter Pro uses the same job description context to generate a tailored cover letter using the Problem-Solution framework — the structure hiring managers actually respond to.
Version History
Job searching means applying to multiple roles. CVwise saves up to 10 CV versions, each linked to the specific job it was optimized for. Users can review, re-download, or build on any previous version without losing their work.
The Business Model
CVwise operates on a freemium model designed to be genuinely useful before you pay anything. The free plan includes 3 AI optimizations per month — enough to test the product properly and see real results. The Pro plan at $11.99/month unlocks unlimited optimizations, Cover Letter Pro, Harvard CV format output, and full version history.
We believe in earning upgrades, not forcing them. Free users get real value. Pro users get everything they need for a serious job search.
A Living Product
CVwise is not a finished product — it's a continuously evolving platform. We update the AI optimization prompts regularly as job market conditions shift, as new ATS systems emerge, and as we collect feedback from real applications. When a new Claude model releases with better instruction-following or reasoning, we upgrade. When users report that a particular industry's CV conventions differ from the defaults, we adjust.
We also continue consulting with HR professionals and career experts to ensure the advice embedded in CVwise reflects current hiring practice — not outdated conventions from five years ago. The job market changes; CVwise changes with it.
If you're in the middle of a job search — or about to start one — try CVwise free. No credit card required. First 3 optimizations on us.