Huntchy is the intelligence layer on top of your ATS. It learns what every client actually accepts — and tells you, before you submit, which candidate closes.
Reads the job description and the CVs. Picks the strongest engineer on paper. Useful — but it's what any recruiter gets free in 30 seconds, and it doesn't know your client.
Reads the CVs plus what this client has actually accepted before — revealed preference, rate ceiling, two-sided fit. The layer an LLM can't have, because it's learned from your outcomes.
Walk in to a prioritized day — what's tight, who's cooling, where to start.
Paste LinkedIn + CV. See who fits this client, with the reasoning and what to confirm.
Close the loop, and a private mirror of how you're growing — yours alone.
Adjectives. "Banking experience probably helps. The rate seems reasonable. Job-hopping may be a concern." Plausible — and the same thing any recruiter already knows.
Probabilities calibrated on real outcomes. Not what might work — what actually closed, measured across the network.
Huntchy assists, it never decides. It shows the fit and what to confirm, never who to cut. Your performance is private to you — never a surveillance tool for your manager. It's that privacy that makes the data honest, and the honest data that makes the matches sharp.
Every staffing agency runs on a quiet truth: the candidate who wins a submission is rarely the best one on paper. It's the one this client accepts — for reasons that never make it into the job description. The domain they won't compromise on. The rate they actually close at. The working style they quietly reject.
Today that knowledge lives in a few senior recruiters' heads. It walks out the door when they leave. New recruiters spend years rebuilding it, one rejection at a time. And no tool captures it — the ATS stores records, not judgment; an LLM reads the CV, not the client.
We built Huntchy to be the memory of the match. It learns what every client truly accepts from real outcomes, and puts that intelligence in front of every recruiter — junior or senior — at the moment they decide who to submit. Not to replace their judgment. To back it.
We're starting narrow on purpose: IT contract staffing in Iberia, where the gap between "best CV" and "right fit" is widest and most expensive. One inch wide, one mile deep.
We cover the four points where the match is decided. The rest stays in your ATS. We're a layer, not a replacement.
Individual performance is private, always. It's not just ethics — it's what makes the data honest enough to be useful.
Huntchy shows the fit and what to confirm. The human submits. Every conclusion shows its source.
The intelligence compounds from your outcomes. No theater, no invented precision — real signal, learned over time.
You already pay €100–180/seat for LinkedIn Recruiter. Huntchy costs a fraction — and tells you which of those candidates actually closes.
How the best submissions actually get made — and why the best CV usually loses.
The strongest CV ranks first and gets rejected at interview. A submission is a bet on six factors at once — and the CV is only one of them. Why the best candidate on paper usually loses, and how to read it.
Job descriptions are wish lists. The rejection pattern is the truth. How to read what a client actually accepts — and why it's the most valuable thing on your desk.
Surveillance tools make recruiters game the data — and poison the very signal they're trying to capture. The case for privacy as a feature, not a compromise.
Anyone can wrap GPT around a job board. What can't be copied with a better prompt is the layer learned from your outcomes. How the six-dimension engine actually works.
Every client has a day-rate they won't cross — and it's rarely the one in the brief. How to read the unwritten ceiling where contract deals actually close.
Recruiter performance is private — always. Hunches, accuracy, the growth mirror, blind-spots: these stay with the recruiter and never roll up to a manager. It's not only ethics. A watched recruiter games the system and poisons the data the whole network depends on. Privacy is what keeps the signal clean.
The system speaks in probabilities, not names. Huntchy never says "Client X rejects 73%." It says "clients of this type are unlikely to accept above this rate." Intelligence is aggregated and anonymized across many clients — no single client or candidate is identifiable in any output. We act on what we learned; we don't expose who we learned it from.
A human confirms every candidate. Before a profile is processed, the recruiter confirms the person is a real candidate for a real role. That confirmation — who, when — is logged. The recruiter and their agency are the data controllers; Huntchy is the processor acting on their instruction.
We minimise by default. For learning and validation, we use anonymized patterns — skills, rate, setup, outcome — not nominal profiles. Personal data is processed only where there's a legal basis and a genuine recruitment process behind it.
Sell your data. Show your numbers to your manager. Name a client or candidate in any aggregated output. Scrape profiles without a lawful, consented source.
Access, correction, deletion, and objection — as required under GDPR. Candidates and clients can request what's held and have it removed.
Processed within the EU where possible, with event-level audit trails. Every decision traces to its source and what was confirmed vs. inferred.
Recruitment AI is treated as high-risk under the EU AI Act. Every dimension is explainable and anchored in a validated model — never a black box.