Crustdata Recruiter

Claude Skills to turn Claude into a 100x Recruiter

Hiring

Artificial Intelligence

Tech

Visit WebsiteSee on Product HuntTwitter ⧉

the data underneath is the real moat here 👏 agencies or in-house biting more?

Congrats on the launch! Turning 1B+ profiles into Claude Skills that explain every candidate they surface is a genuinely useful trust layer, vs most other recruiting AI just gives you a ranked list and a shrug. Once a recruiter overrides a pick a few times, does the skill actually learn from that feedback, or does every search start from the same baseline judgment?

Very interesting and potentially game changing.

As a recruiter in another life, one issue we would run into is out of date resumes, linkedin profiles, etc. Does Crustdata have a way to cross reference resumes with current jobs, resumes on other job boards, or some way to validate that the candidates profile is up to date?

@mhimed_crustdata Just voted, congrats. Claims like "100x" need fast trust-building right after signup, curious what that first email looks like.

I wonder if this works best for recruiters who already know exactly what they're looking for. If the hiring brief changes every week, does the learning still compound?

@Nithish A that's great, having it explain itself rather than just being a black box is honestly the difference between me trusting it and just double checking everything it does anyway. Good to hear it's already there at v1.

been running crustdata daily across concurrent searches in robotics, autonomy, defense, semi, and sensors, and the thing i keep coming back to is how fast it takes me from a fuzzy brief to a signal-heavy shortlist. between pre-checking filters before spending credits, reading employer + career arc + title-shape in one pass, and closing personal-email gaps other tools just won't touch, i'm getting to the right people before other desks even finish their booleans. the more i use it, the better it reads how i actually think about a candidate.

the speed isn't even the point though. because i'm not sinking hours into profiles that were never going to hit, i get to spend real time on the relationships that actually move the needle. the long-arc, stay-in-touch-through-three-job-changes kind that used to be the whole job before templated cold email flattened everything. right now most engineers are getting 20 near-identical emails a week from people who don't know the space, and the bar to call yourself a sourcer has never been lower. i feel like crustdata has been the way for me to get back to recruiting the way it worked before all this noise: fewer, sharper conversations, sent to the right inbox, compounding over time.

huge congrats team.

We've been using Crustdata for quite a while now, and honestly they've been one of the best vendors we've worked with.

Really like this direction too. I think the interesting part isn't just searching 1B+ profiles—it's that the system learns how you evaluate candidates over time instead of making you start from scratch on every search. The explainability is also a nice touch. As AI gets more involved in recruiting, understanding why a candidate was recommended becomes just as important as the recommendation itself.

Also just wanted to give a shoutout to the team. Every time we've had questions, missing data, or feature requests, they've been incredibly responsive and thoughtful. Great product, even better people.

Congrats on the launch! 🚀

@Mhimed thanks for the reply - good to hear it's feedback-driven rather than needing a big volume of searches first. that actually shifts my earlier concern a bit: if it adapts fast off explicit feedback, is there a way to see or edit what it's "learned" so far, like a running summary of preferences, or is it more opaque and you just keep correcting as you go?

is there a way to test it on one real role before committing? I have an open req where I already know the good candidates, if your list finds them plus some I missed I’ll be convinced

Does this run in normal claude.ai or do I need claude code? My sourcing team is not technical so curious what they should end up using?

Can it monitor a search over time? I have 2-3 roles that are basically always open, would love to get notified when someone matching changes jobs instead of re-running the search every week

the part that gives me pause is the opposite side of the personalization pitch. if it saves rulings like 'never anyone from big aerospace' as plain-language feedback and applies them going forward, some of that feedback is going to be legitimate judgement and some of it is going to be a recruiter's unexamined bias, and the tool has no way to tell those apart. a human recruiter's bias is inconsistent and limited in scale, this would encode it and run it the same way every time. is there any check on what kind of rulings get accepted, or does it save whatever plain-english correction it's given

Congrats on launching! which ATS integrations are live right now? we're on ashby and the push-to-ATS part is very challenging right now when I work with other tools

Congrats! how does it work with multiple recruiters in one team? we're 5 and we definitely don't rank candidates the same way. does everyone train their own version?

@nithish_a1 Does it also give contact info (emails and phone numbers)? We are using rocketreach + our sourcing tool for that and if I somehow can replace both with Claude and your MCP that could be very interesting

I’m not a recruiter, but I can immediately see how much time this could save. Filtering through endless profiles sounds simple until you’ve done it for hours and still don’t get the people you actually want. The fact that it learns what the recruiter values makes this much more useful than just another search tool. Congrats on the launch!!👏

What about candidates who are not active on linkedin? recruiting a lot of engineers at the moment and not all of them have an active Linkedin profile so how are you handling that? And congrats on launching!