Hi HN, We’ve been building a hiring tool that ranks candidates using multiple job-specific criteria rather than keyword matching (traditional ATS-style filtering).

For each job listing, the system generates hiring criteria and sub-criteria, then analyzes and ranks candidate CVs based on how well they meet those criteria. The original target users are hiring agencies, HR teams, and recruiters.

Our current challenge is early adoption. Agencies are hard to onboard without seeing real results, and real results usually require usage at some scale.

So we’re running an experiment: We added a candidate-facing tool where job seekers can paste a job link, upload their CV, and get an analysis showing which hiring criteria they meet and which they’re missing.

The hypothesis is that candidate usage and data can act as a bottom-up wedge to eventually attract agencies and recruiters.

We’re early and unsure whether this approach will work but let's see.