TalentElse
30,000+ users, 5,000+ resumes processed
The problem
Recruiters were reading resumes one by one and tracking candidates across spreadsheets and inboxes. Good candidates were slipping through the gaps.
What I built
I built the platform end to end: resume parsing, an ATS scoring pipeline, candidate pipelines and a recruiter dashboard, deployed on a managed Postgres backend.
Key decisions
Score with evidence
Every match score shows the parts of the resume that produced it, so a recruiter can trust or overrule it.
Queue the heavy work
Parsing runs in background jobs with retries, so uploads never block the interface.
Row-level security
Each company only ever sees its own candidates, enforced in the database, not just the app.
Own the data model
A plain relational schema keeps reporting and future integrations simple.
Results
30,000+
users
5,000+
resumes processed
Need something like this?
