What happened to Proxycurl

Proxycurl is gone. Visit nubela.co/proxycurl today and you are met with an interstitial that says, in the company's own words, "Proxycurl is no longer in service. See NinjaPear, a data platform for customer data instead." followed by "We've Moved! The founder of Proxycurl is now working on NinjaPear, a B2B Customer Data Platform. Redirecting in 3 seconds..."

The timeline is on the public record, and unusually for this category, every step of it can be checked:

DateWhat happenedWhere it is recorded
24 January 2025LinkedIn Corporation v. Nubela Pte. Ltd., Proxycurl LLC, Steven Goh, Bach Le, No. 3:25-cv-00828, Northern District of California. The complaint pleads breach of contract, fraud and deceit, the Computer Fraud and Abuse Act, and unfair competitionFederal docket, entry 1
27 January 2025Summons issued as to all four defendantsFederal docket, entry 8
4 July 2025The founder announces the shutdown: "In January earlier this year (2025), LinkedIn filed a lawsuit against Proxycurl." He describes it as a "~$10M revenue business before we had to shut it down"nubela.co/blog
25 July 2025Judgment enteredFederal docket, entry 28
29 April 2026The founder writes: "Even though we settled, we do not want customers to carry a legal risk forward"nubela.co/blog

One deliberate omission. We are not going to tell you what the July 2025 judgment says. The docket entry is titled "Judgment" and we have not read the document itself, so any description of its terms would be us guessing at a court order in public. Several pages about this case do exactly that. The date and the founder's own word, "settled", are what the record supports, and that is all we will state.

Why the no-code question is different

Search for a Proxycurl replacement and you will find pages written for engineers: endpoint mappings, rate limits, response schemas, migration snippets. That is reasonable, because Proxycurl was an API and most of its customers wrote code against it.

But a large share of the people who relied on it never wrote a line. They wired it into Zapier or Make, or into a Google Sheet, and what they lost was not an endpoint. It was a working step in the middle of something else. Almost nobody is writing for them, so here is that page.

The honest starting point is that an API and an extractor are not the same product, and swapping one for the other changes how you work. An API enriches a record you already have, on demand, invisibly, at whatever scale you can pay for. An extractor gives you a list from a search you ran yourself. If your automation depended on enriching arbitrary profile URLs at volume without a human present, no browser tool replaces that, and you should read the developer pages after all.

No-code options after Proxycurl, September 2026

ToolShapeNeeds codeFree tierBest for
LeadsforlinkedChrome extension plus web appNo100 leads, no cardTurning a search, a post's commenters or an event's attendees into a CSV
NinjaPearCustomer data platform, the founder's own successorYes for the API, and it is the direct lineageNot confirmedTeams replacing the original endpoints
PhantomBusterCloud automation platform, 130+ ready-made "Phantoms"No, but there is real setupYes, and crippled: 30 minutes a month, one slot, and a 10-row export capChaining steps without writing code
ClaySpreadsheet-shaped enrichment with many data sourcesNo for basic useNot on the current self-serve tiersWaterfall enrichment across providers. Repriced 11 March 2026
Captain DataAPI-first data platformYes, nowNoNothing, if you are avoiding code. Its own changelog dates the no-code v3 sunset to 9 January 2026
Lobstr.ioHosted scrapers with schedulingNoYesRecurring scheduled collection
ScrapingdogScraping APIYesTrialDevelopers, not the audience of this page

Read from each vendor's own pages on 5 September 2026. Note the pattern in the "needs code" column: the no-code middle of this market got thinner in 2026, not wider. Captain Data left it deliberately.

What the founder himself recommends

It is worth saying plainly, because it is the most direct evidence available and it is not ours: the person who built Proxycurl points people at NinjaPear, his new B2B customer data platform, and writes that "NinjaPear is what comes after Proxycurl." He maps the old endpoints to new ones, an Employee API for person enrichment, a Company API for company data and a Monitor API for signals. If you were an API customer, start there rather than with us.

The nubela.co navigation also still lists Sapiengraph, described as "B2B Enrichment in Google Sheets", which on the face of it is exactly the no-code shape this page is about. We could not confirm that it is still sold, or what it now sources its data from, so we are not going to recommend it. Check it yourself before relying on it. That gap is precisely the kind of thing that ends up costing people a workflow twice.

What to check before you pick the next one

The useful lesson from this shutdown is not which tool to buy. It is what to ask about whatever you buy next, because the people hurt most by Proxycurl closing were the ones with no copy of their own data and no way to get one.

  • Can you export everything, on the plan you are actually on? Not on an upgrade. If the answer is no, you are renting your data.
  • What happens to your data when you stop paying? Almost nobody in this category documents this. It is worth asking in writing before you subscribe.
  • Where does the data physically come from? A vendor holding a bulk scraped database carries a different legal profile from a tool that reads pages in your own browser session as you open them. Proxycurl's LinkDB was the former.
  • Is there a person behind the product who answers? Proxycurl's founder spent months, in his words, "assisting our existing customers with deboarding Proxycurl." That is more than most shutdowns offer.

We do not claim to be immune to any of this. Every tool that touches LinkedIn operates on a platform whose rules it does not set, and nobody can promise otherwise. What we can do is make the data yours: CSV export is included on the free plan, so you can take everything you have collected out of our product at any time, without upgrading and without asking.

Which one fits which job

If you wrote code against Proxycurl, read the founder's own migration notes and look at NinjaPear. We are not an API company and we would be a poor answer to your question.

If you used it inside Zapier, Make or a spreadsheet, decide first whether you need enrichment of records you already hold, or discovery of people you do not. For enrichment, look at Clay. For discovery, an extractor is simpler and cheaper.

If what you really needed was a list of people from LinkedIn, that is our job. A search, a post's commenters, a post's reactors, an event's attendees, a group's members or a company's people page becomes a CSV with name, headline, job title, company, location and profile URL. 100 leads free on signup, no card, released in batches of 20. After that, packs from EUR 1.99 for 500 leads that never expire, or unlimited scraping on Basic at EUR 4.99 a month, or EUR 2.99 a month billed yearly. One credit is one lead and duplicates are refunded. Sales Navigator is optional.

Frequently asked questions

Is Proxycurl still available in 2026?

No. The Proxycurl page on nubela.co now shows an interstitial reading 'Proxycurl is no longer in service. See NinjaPear, a data platform for customer data instead.' The founder announced the shutdown on 4 July 2025. An old trial button still renders behind the notice, but the API is not purchasable and should not be treated as one.

Why did Proxycurl shut down?

LinkedIn filed suit on 24 January 2025 in the Northern District of California, case number 3:25-cv-00828, against Nubela Pte. Ltd., Proxycurl LLC and two individuals, pleading breach of contract, fraud and deceit, the Computer Fraud and Abuse Act and unfair competition. The founder wrote on 4 July 2025 that this was why the company closed, describing it as a roughly 10 million dollar revenue business. Judgment was entered on 25 July 2025 and the founder later referred to having settled. We have not read the judgment document, so we do not state its terms.

What is the best Proxycurl alternative if I cannot write code?

It depends on which half of the job you need. For enriching records you already hold, Clay is the closest no-code equivalent. For collecting people from LinkedIn in the first place, an extractor is simpler and cheaper, and Leadsforlinked does that with a Chrome extension and a CSV export included on the free plan. If you need arbitrary profile enrichment at API scale without a human present, no browser-based tool replaces that and you should look at the developer options instead.

Did Proxycurl customers get into trouble themselves?

There is no evidence of that and we are not going to suggest it. The lawsuit named the company and two individuals connected to it. What the founder did say publicly, in April 2026, is that even after settling he did not want customers to carry legal risk forward, which is why the service was not transferred to anyone else.

What should I ask a data vendor before I rely on it?

Four things, in writing. Can you export everything on the plan you are actually on, rather than after an upgrade. What happens to your data when you stop paying. Where the data physically comes from, since a bulk scraped database carries a different profile from a tool reading pages in your own browser session. And whether a real person answers when something breaks.