Live Company and People Data via APIs Search, enrich, or watch a person or company in real-time. Get instant updates when an event happens.
For example, get updates when a person a) changes job or b) gets a promotion or c) adds a new skill or d) makes a new post.
This is the 5th launch from Crustdata. View more
Crustdata Recruiter
Launching today
A set of recruiting skills that runs inside Claude on live data from 1B+ profiles from Crustdata. It learns your judgement with every search, ranks candidates the way you would, explains every candidate it picks, and drafts outreach into your ATS.
Launch Team
Crustdata
This looks like a great tool - congratulations on the launch. 1 billion profiles is pretty impressive! My question is primarily about the way the data is harnessed and the sustainability of that model. Put simply, if you are pulling candidate data from a job board then you are potentially skipping the need for a recruiter to have a licence for that job board? In which case, there is a future risk to the job board's sustainability commercially. And then, of course, you lose that data source. Is that something you've got covered? If so, how do you compensate your data sources?
@martin_tanner Hey Martin, thanks for the thoughtful question! Our candidate data comes from the open web, places where people publicly list their own work history, projects, research and so on.
the "learns your judgement with every search" line is the part I'd want to poke at. how many searches does it actually need before the ranking starts feeling like yours instead of a generic Claude default, and does that learning carry over if two recruiters on the same team have genuinely different taste for the same role? feels like it could either be the killer feature or a slow-to-converge black box depending on the answer.
@omri_ben_shoham1 Hey Omri, great question! From what we've seen, it starts feeling personal within the first few searches, but it really depends on the feedback, if you tell it why someone is or isn't a fit, it will adapt very fast with next search.
honestly this looks really useful, one thing i'd love is a way to test the api in a sandbox with fake data before committing. like a playground where you can see what job change or promotion alerts actually look like in the response. would save a lot of trial and error
Crustdata
@ferdi788603 That's a great idea Ferdi, thank you! Will pass this on to the team!
@ferdi788603 Hey Ferdi, this is now possible! You can get an API key with free credits and test out the entire workflow without having to commit to anything!
This is impressive! Curious, how quickly does the model adapt when a recruiter's judgement shifts for a new role with different requirements than previous searches?
Best of luck with the launch excited to see how this reshapes sourcing workflows!
Crustdata
@aymi_malik Hey Muhammad, from our experience it does a pretty good job of adapting to different roles with different requirements, but the quality of it lies in the feedback the recruiter provides to the models. The better the feedback, the quicker and the better results they get!
@aymi_malik @nithish_a1 Hey Muhammad, adding to Nithish's point, it will keep getting better with every new search you make, since Claude keeps the context from your previous searches saved in the skill.
@nithish_a1 Whats the data source for candidates? outdated profiles have always felt like a flaw in the industry you just live with. how do you guys keep it fresh?
Crustdata
@maurya_abhiranjan We get data from across the web for candidates. Any place where folks list their work history, their work, research papers - we get it from there. And we keep pulling this data from the web in real-time, hence the candidate data is always accurate.
@maurya_abhiranjan @nithish_a1 Adding to what Nithish said, we also detect any change events (job switches, promotions, title changes) in real-time, so you can search but also filter by candidates who just joined or just left a company for example.
How long before it actually starts thinking like me? like how many rounds of feedback until it's calibrated?
@suyash_kr Hey Suyash, there's no fixed number! It gets better with every search since Claude keeps the context. The better the feedback you give it, the faster it calibrates!