Brief to shortlist — AI candidate matching for a high-ticket placement firm

A high-ticket executive search firm places senior people into roles where the wrong hire is expensive in every way. Their value to clients is the depth of the relationship and the quality of the match — but their candidate intelligence lived across a dozen tools, half a dozen recruiters’ heads, and a long history of side notes from past searches. Every new search started by reassembling the picture.

What we built

A candidate intelligence system that keeps the depth without keeping the chaos. Every candidate has one profile — career history, expertise, references, recruiter notes from every interaction. Every client brief is captured the same way. AI matching reads the brief and surfaces candidates from the database whose profile actually fits the role, ranked with the reasoning visible — so recruiters can edit the shortlist with their own judgment, not against an opaque score. The relationship stays human; the lookup gets fast.

How we built it

A Next.js dashboard on Vercel, Supabase as the source of truth for candidates, briefs, searches, and recruiter notes. OpenAI handles the matching — reading each brief and each candidate profile, returning a ranked shortlist with explanations the recruiter can read and override. The reporting layer shows search progress and historical placement quality, so the firm can sharpen its own playbook over time.

The outcome

A firm that places senior people faster without losing the quality of the match. The recruiters keep doing what their clients pay them for — building relationships, sharpening briefs, judging fit — and stop spending half their week on the lookup that should have been instant.