Eventually I realized the problem wasn't the resume quality. It was that ATS systems were filtering applications out before a human ever read them. Most job seekers have no visibility into why this happens or how to fix it per role.
So I built CareerKit.
You upload a resume and paste a job description. It returns an ATS compatibility score, flags missing keywords and skill mismatches, then generates a tailored resume and cover letter for that specific role.
Tech stack: React, TypeScript, Node.js, Python.
The target user is early-career professionals and recent grads. They tend to have less natural keyword overlap with job descriptions because they have shorter work histories, which makes ATS filtering hit them harder than experienced candidates.
Free tier available with no card required.
A few things I am uncertain about and genuinely want input on:
1. ATS scoring feels like the right hook but I keep wondering if the resume generator is actually more valuable to users. Hard to tell from early usage. 2. The free tier is limited (1 resume analysis, 1 resume generate,3 cover letters lifetime). I am not sure if that is tight enough to convert or too stingy to show value. 3. Keyword matching currently works by comparing the resume text against the job description. Curious if anyone has thoughts on better approaches here.
Happy to answer questions about how it works.