The Problem
Generic resumes don't work. Job seekers need resumes tailored per application, but manual tailoring is slow and inconsistent. Existing tools either lack AI intelligence or lock users into specific providers with privacy concerns.
Research & Insights
Studied resume tailoring workflows of job seekers. Identified that existing tools either lack AI intelligence or lock users into specific providers with privacy concerns.
Design Process
Designed profile-first architecture — users build a master profile, then AI generates tailored versions per job. Built provider-agnostic AI routing supporting OpenRouter, Anthropic, and OpenAI. Implemented multi-format export (PDF, DOCX, Markdown).
The Solution
Self-hosted, privacy-first resume tailoring tool with master profile management, job description parsing, AI-powered tailoring, and multi-format export. Admin system for managing AI providers and configurations.
Results & Impact
Active and functional. Demonstrates full-stack product thinking from UX research through to deployed AI product with real users.
What I Learned
Building a tool I use daily sharpened my feel for AI UX. The handoff between human judgment and AI output is where most products fail, and it can't be solved by making the AI more "helpful."