Hi Product Hunt! 👋 I built CVShift to solve a massive, silent problem in the tech industry: "Beautiful" resumes are getting highly qualified people auto-rejected. Most popular resume builders on the market prioritize aesthetics over data structure. Under the hood, they use client-side libraries (like html2canvas) to take a screenshot of your CV and save it as a PDF. When a recruiter’s Applicant Tracking System (ATS) scans that file, it reads absolutely zero text. The algorithm assumes you have no skills and instantly drops you from the pipeline. I got fed up with this and engineered CVShift to prioritize machine-readability first. Here is what makes CVShift different: • True Document Compilation: We offloaded document generation to Supabase Edge Functions (Deno). Instead of flattening your CV into an image, we programmatically compile strict, native DOCX and PDF files that ATS parsers can read perfectly. • Edge NLP Parsing: Importing existing CVs (like massive LinkedIn exports) used to crash browser memory stacks. We built a custom chunking algorithm that streams your data to the Edge for intelligent NLP parsing without locking up your browser. • Geo-Targeted Job Board: We integrated a background engine using Cloudflare D1 that aggregates fresh jobs and serves them to you dynamically based on your IP region, dropping anything older than 30 days. The ultimate goal: We are experimenting with embedding W3C Verifiable Credentials directly into exported CVs so next-gen ATS software can cryptographically verify employment history in milliseconds. The core builder is completely free to use. I would absolutely love for the PH community to stress-test the PDF importer, generate a CV, and let me know how the ATS scanner performs for your specific industry! Happy to answer any questions about the React/Next.js split architecture, Edge functions, or ATS algorithms in the comments! 👇

CVShift

The ATS-proof resume builder that actually gets you hired.

The ATS-proof resume builder that actually gets you hired.