personal_project
archiveddiliResumerip 💀
Everything a job seeker needs in one place — CV tailoring, cover letters, interview prep, and application tracking.
// what problem this solves
Job hunting is repetitive, but candidate positioning is dynamic. Tailoring a CV, writing a cover letter, preparing for interviews, and tracking applications usually happens across disconnected tools and endless copy-paste workflows. Every step starts from scratch, which means context gets lost, effort gets duplicated, and the search becomes harder to sustain.
// what I built
I built diliResume as an end-to-end AI career workflow engine. Users could upload or create a resume, map it against a job description, generate tailored cover letters, practise interviews with AI feedback, and manage their application pipeline in one product. Instead of solving one narrow problem, the goal was to connect the entire candidate journey into a single continuous workflow.
// how it works
Under the hood, the product ran on a workflow-oriented architecture built with Next.js, TypeScript, and Firebase. Resume and job-description data moved through async processing stages including PDF parsing, content extraction, JD analysis, resume matching, cover-letter generation, and interview-question generation. I implemented status-driven background jobs, retry-aware flows, and a shared data model so resumes, job assets, and interview sessions stayed connected instead of living in separate feature silos.
// result
- Built resume flows for blank CV creation, existing CV reuse, and PDF import
- Shipped JD analysis and resume-match scoring as connected workflow steps
- Generated tailored cover letters from both uploaded PDFs and in-product resumes
- Delivered interview setup, question generation, and AI answer-feedback flows
- Turned scattered job-search tasks into one integrated end-to-end product experience
the stack