AlignCV

AlignCV helps job seekers turn the CV they already have into a version tailored to a specific job description, downloadable as an ATS-friendly DOCX or PDF. It is in private beta.

General-purpose AI tools happily blend a job description into someone's work history, producing CVs that read well but claim experience the person doesn't have. The product had to tailor a CV without ever inventing anything — and prove it, rather than just promise it.

• Product concept, design and the full build, front to back
• CV parsing for PDF, DOCX and plain text into structured sections
• A structured match score with documented weights, so every point maps to a specific matched or missing requirement
• A provenance check: each rewritten line carries a reference to where it came from on the original CV, and anything that can't be traced is discarded before the user sees it
• DOCX and PDF generation that keeps the user's own section order in a single-column, ATS-safe layout
• Google sign-in, quotas, an admin area, audit logging, and optional bring-your-own AI key encrypted at rest

• TypeScript monorepo with a shared package for types and validation used by both the API and the web app
• NestJS API with PostgreSQL via MikroORM, organised into modules for auth, CVs, matching, tailoring, documents, quota and audit
• Long-running work (parsing, generation) handled by background jobs on Redis and BullMQ, so the interface stays responsive
• Uploaded and generated files stored in S3
• Next.js and React front end with TanStack Query

TechnologiesTypeScript, Next.js, React, NestJS, PostgreSQL, MikroORM, Redis, BullMQ, AWS S3, Tailwind CSS, Zod