Everything between apply and offer.
The candidate applies on your job board. The AI handles every step until your team has a scored interview to review. Here's what that looks like.
Customizable, clinically-grounded interviews.
Job board
Candidates apply through your Munchkins job board — a branded, mobile-friendly page you can link from your careers site or share directly. Every application lands in one place, parsed and ready for the next step. No inboxes. No spreadsheets.
You control the listing: role title, shift type, location, and the screening criteria you care about. Munchkins builds the interview rubric from what you give us. The job board is the starting line — everything after it is automated.
Resume parsing
Every résumé that comes in gets read by an AI tuned for clinical roles. It pulls out what actually matters — job titles, employment dates, certifications listed, and gaps. Not a keyword match. A structured read.
The output is a clean candidate profile: role history in order, red flags surfaced, and a note on anything that looks missing for the position. Your team never opens a raw PDF unless they want to.
Automated invite
Qualified candidates get an interview invitation automatically — no coordinator needed to send it. The message goes out as soon as the résumé clears the initial screen, which means candidates hear back in minutes, not days.
The invite explains what to expect: a 15-minute voice conversation with an AI agent, at a time that works for them. Consent is captured before the call begins. Candidates can opt out and request a human interviewer at any point.
AI voice interview
Our voice agent calls the candidate and conducts a real conversation — not a chatbot, not a form. It asks the questions your rubric specifies, listens to the answers, and follows up when something needs clarifying.
The interview runs 15 minutes on average. It covers clinical background, role-specific questions, shift flexibility, and soft-skill signals. The whole thing is recorded. Nothing is lost.
Transcription
Every call is transcribed in full via Deepgram. The transcript is timestamped, speaker-labeled, and stored alongside the original audio. You can read it, search it, or share it with a hiring manager who wasn’t on the call.
The audio and transcript together are the record. If a decision is ever questioned, the trail is there. No notes-from-memory, no “I think they said.”
Structured scoring
An LLM scores the transcript against your rubric — the same rubric, applied the same way, to every candidate. The scorecard shows how each answer mapped to each criterion, with a short rationale for every score.
The point isn’t to replace your judgment. It’s to make sure every candidate gets a fair first look, and your team spends their time on the ones worth a second conversation.
Automated follow-ups
After every interview, follow-up messages go out automatically. Candidates get a status update. Your team gets a notification. Nobody falls through the cracks because someone forgot to send an email.
The follow-up cadence is configurable. You decide what goes out at each stage — we make sure it actually goes out.
See it run on
your hardest role.
See the full flow run on your own job posting. Fifteen-minute demo, no slides.
Patients deserve clinical knowledge.
No commitment · No credit card · 15 minutes