Hi, Does anyone understand how Claygent visits Linkedin profiles. Linkedin is very harsh on scrapers, and with every scraper I've tried, I eventually hit a roadblock. How does Claygent do it?
Yes. Not many people know this, but… Claygent doesn't visit LinkedIn profiles live. It pulls MixRank data via a Langchain tool-call. So, in many cases, when you request LinkedIn data via Claygent (w/ an API key) it's cheaper than spending Claygent credits on their LinkedIn enrichment column directly (which also calls MixRank). Hope this helps!
so what's the best way to get live linkedin data from a profile?
"the data clay provides is pretty darn close to live" it really depends. i worked with a customer where some prospects changed jobs two years ago and weren't updated in clay data, so i use "enrich person" to get the live data.
Right, the first part of your message refers to “Find People” data, which is bad >50% of the time. The second part of your message refers to “Enrich Person”, which is the thing I've been talking about this whole time. Clay uses MixRank to fuel their “live data” enrichment (aka “Enrich Person”). And it's the same source used in a tool-call by Claygent when LinkedIn data is requested.
Elric L. I've tried using claude code with browser harness to scrape linkedin it has worked for me but it's slow because it literally opens linkedin in your browser and scans through the profile
i wouldn't use that; you can get your account blocked/banned. i got blocked already doing that.
I don't use MixRank directly. I use Clay’s MixRank (aka “Enrich Person”) plus a custom “contact washing machine” workflow that flags the bad contacts
Elric L. ohh i see, anyways I use it on my secondary account so I can experiment with it
