About the project
We turned ambiguous profiles into actionable signals, tailored to recruiter priorities.
Primary Challenges:
We added more humanity to the automated recruiting process.
Our team began by clearly defining the problem: in a sea of identical resumes, how does a recruiter choose which entry-level applicant to interview?
We interviewed recruiters and hiring managers to pinpoint key pain points in the selection process, then supplemented those insights with focused secondary research.
Research Highlights
Our core concept wasn’t to reinvent the job board, but to enhance it — giving hiring managers a smarter way to evaluate entry-level candidates who are otherwise indistinguishable.
Our prototype analyzed extracurriculars, prior interests, and location-based context to infer job success based on previous hiring outcomes — modified by recruiter input. By training on preferences and results, the system learned to highlight candidates likely to succeed — even when their resumes appeared otherwise unremarkable.
Candidate list visualizes non-standard signals as ranked, decision-ready data.
Individual reports surface transferable strengths and recruiter-aligned fit.
Job descriptions enhanced with AI-inferred success signals from prior hires.
Recruiters define their criteria and signal weighting — training the system in real time.
