HR Tech · AI-assisted matching demo · Global demo
A demo app that uses Claude to score a candidate profile against a job description — surfacing match (0-100%), missing skills, and a one-paragraph rationale. Built to show that AI-assisted screening works before a company commits to a recruiter.
2025
HR teams spend hours screening resumes for one role. AI-assisted screening exists but most products require a recruiter to upload the candidate, configure the role, and parse the result — three steps before the value shows. A demo app that runs the whole flow in under 30 seconds, with no setup, is the only way to show the value to a non-technical buyer.
Carmen is a single-page demo. The buyer pastes a job description and a candidate resume, hits Analyze, and within 30 seconds gets a match percentage (0-100%), a list of missing skills, and a one-paragraph rationale that quotes the resume. No login, no upload UI, no settings page. The prompt is engineered to force structured JSON output so the UI can render scores consistently. Behind the scenes it uses Claude via the AI SDK on Vercel; the whole demo deploys as a single Next.js app with one serverless route.
Independent practice based in Cali. Automation flows, hand-built UI, and end-to-end web — for teams that would rather have it done than debate it.
Let's talkFrom intake to deploy, the studio keeps work in one place: scope, design, build, ship. Briefs land on the desk, deliverables land on the client.
Andrés Morales
