> Index of everything AyHire publishes for agents: https://ayhire.com/llms.txt

# Firecrawl Directory vs AyHire for hiring engineers

https://ayhire.com/blog/firecrawl-directory · 28 October 2026 · 9 min read · Nawneet Kumar

> Firecrawl Directory puts 240M+ profiles in ChatGPT and Claude. What it costs, what a profile can't show about an engineer, and when AyHire fits better.

Firecrawl launched **Firecrawl Directory** on 28 October 2026: "free recruiting data for ChatGPT & Claude". You describe a role in a chat, and your agent builds a shortlist from more than 240 million professional profiles, with job histories and work emails. Searching is free, and every account gets 1,000 credits a month for the details.

If you hire engineers and already work in Claude or ChatGPT, it is worth a look. It is also a good moment to ask what a profile can tell you about an engineer, and what it can't.

This post covers what Firecrawl Directory does, what it really costs, the same search run from engineers' public work instead, and when each tool is the better fit. AyHire is our product, so read those parts with that in mind. Every Firecrawl claim comes from Firecrawl's own pages, opened on 28 October 2026.

## What is Firecrawl Directory?

Directory is a new collection in Alexandria, Firecrawl's library of data providers for AI agents. Firecrawl's [launch post](https://www.firecrawl.dev/blog/introducing-directory) describes it like this:

- **The data**: "more than 240 million professional profiles, including employment histories and work contact information". The launch tweet says 250M+.
- **Where it comes from**: providers "including Apollo, FullEnrich, and Data Legion".
- **How you search**: by job title, company, location and keywords. You can also paste a job description and ask your agent to build a shortlist.
- **What comes back**: "available information about someone's current role, previous employers, professional background, and work contact details".
- **Where it runs**: in ChatGPT and Claude through the Firecrawl plugin, and in terminal agents through Firecrawl's skills.

The launch post's own example: a search for senior backend engineers in New York "returned 142 matches in a single free search."

## What does Firecrawl Directory cost?

Searching is free. Credits are spent when your agent pulls the details, contact details included, and every account gets 1,000 free credits a month with no card.

Firecrawl's [post on people search in Claude](https://www.firecrawl.dev/blog/claude-for-people-search) lists the credit prices as of 18 October 2026:

- **Apollo people search**: 0 credits.
- **Apollo person match** (work email, title and job history): 30 credits per matched person.
- **FullEnrich people search**: 5 credits per profile returned.
- **FullEnrich work email**: 20 credits per email found.
- **FullEnrich mobile number**: 175 credits per number found.

By Firecrawl's own count, 1,000 credits covers "about 30 Apollo matches or 50 FullEnrich work emails". The free plan stops with an error when the credits run out, so it never charges you by surprise. Past that, [Firecrawl's paid plans](https://www.firecrawl.dev/pricing) start at $19 a month for 5,000 credits.

For a founder writing to a few dozen people a month, that is close to free.

## What a profile can't tell you about an engineer

Firecrawl says this part plainly. "A profile is a useful starting point, but it doesn't tell the whole story," the launch post says, and Directory is "not a replacement for talking to candidates or evaluating their work."

That matters more for engineers than for most roles. A profile gives you titles, employers and dates. Two people with the same title at the same company can have done very different jobs: one designed the storage layer, the other kept an internal dashboard running. A keyword search for "infrastructure at scale" finds the people who wrote those words on their profile, which is not the same as the people who built it.

What tells you whether someone can do the job is the work: the commits they wrote, the pull requests that got merged, the reviews they gave, the READMEs that explain what they built. For many engineers, some of that is public on GitHub.

Firecrawl's answer is a second step. Once you have a name, your agent can run a web search for "open-source contributions" and other public work. That gets you links, one person at a time. Reading them, and judging them against your role, is still your job.

## The same search, starting from the work

Firecrawl's suggested prompt is "Find senior backend engineers in New York who've worked on infrastructure at scale." On 28 October 2026 we ran the work-first version through AyHire's MCP server. Instead of searching titles, it searched GitHub for the infrastructure itself: repos tagged `distributed-systems`, written in Go, with 1,000 stars or more, pushed since 1 September 2026.

```
topic:distributed-systems language:"Go" stars:>=1000 pushed:>=2026-09-01
```

31 repos matched. AyHire read the 8 most-starred and took the five people who wrote the most of each: 40 people, each with the repo they were found through and their rank in it. The search reads GitHub only, so it cost nothing. Here are the projects, by the commits of their top five people:

- [etcd-io/etcd](https://github.com/etcd-io/etcd/graphs/contributors), 52,432 stars: 12,101 commits.
- [seaweedfs/seaweedfs](https://github.com/seaweedfs/seaweedfs/graphs/contributors), 35,794 stars: 11,470 commits, 10,791 from one person.
- [dapr/dapr](https://github.com/dapr/dapr/graphs/contributors), 26,166 stars: 3,463 commits.
- [temporalio/temporal](https://github.com/temporalio/temporal/graphs/contributors), 24,045 stars: 3,723 commits, spread evenly (524 to 935 each).
- [nats-io/nats-server](https://github.com/nats-io/nats-server/graphs/contributors), 20,954 stars: 11,270 commits.
- [rqlite/rqlite](https://github.com/rqlite/rqlite/graphs/contributors), 17,878 stars: 8,593 commits, 8,548 from one person.
- [juicedata/juicefs](https://github.com/juicedata/juicefs/graphs/contributors), 14,585 stars: 2,708 commits.
- [encoredev/encore](https://github.com/encoredev/encore/graphs/contributors), 12,650 stars: 1,904 commits.

Every one of the 40 is tied to a project you can open and to their share of it, and two of these projects are almost entirely one person's work. 7 of the 40 have GitHub's "available for hire" flag on. None of that depends on what anyone wrote about themselves.

There is a fair point the other way. None of the 40 lists New York on GitHub, and 10 list no place at all. If the city is a hard requirement, a profile database will find you more people in it. AyHire can keep people whose GitHub profile names a place, but it never scores anyone on where they live.

And 40 strong builders is still not a decision. Which of them worked on consensus, and which on the client libraries? Who reviews other people's code? AyHire's next step answers that: it reads each person's commits, pull requests and READMEs against your role's criteria and links what it found. We list projects here, not people: the developers didn't ask to be on a marketing page. To do the same search by hand, see our guide to [finding engineers on GitHub](/blog/find-engineers-on-github).

## Firecrawl Directory vs AyHire

- **Where people come from**: Directory searches 240M+ profiles from Apollo, FullEnrich and Data Legion, for any role. AyHire starts from GitHub: it finds the repos that match the role, then the people who wrote most of them.
- **What you get back**: Directory gives profiles (current role, past employers, background and contact details) for your agent to sort and compare. AyHire gives a call on each person for your role: who to talk to first, one line of why and one thing to check, with a link to the commit, pull request or README behind each judgement.
- **How fit is judged**: Directory leaves it to you and your agent; in Firecrawl's words, "the goal isn't to automate hiring decisions." AyHire judges each person against your role's criteria, each marked Must have, Important or Nice to have, and an unknown never counts against anyone.
- **Their work**: Directory reaches it through a separate web search, one person at a time. AyHire reads public repos, commits, pull requests, READMEs and a year of work on other people's projects.
- **Contact details**: Directory finds work emails and mobile numbers, paid by the credit. AyHire finds LinkedIn, a work email and X, each with its source and how sure it is, and says whether someone looks open to work. Those lookups come with the plan; there are no credits.
- **Where it runs**: both work inside Claude and ChatGPT. AyHire also connects to Claude Code, Codex, Cursor and other coding agents over MCP, and has a web app with roles, a board and a shared shortlist.
- **ATS**: Firecrawl's launch post doesn't mention one. AyHire sends shortlisted people to Greenhouse, Lever and 11 more ATSs, or to any tool through a webhook.
- **Price**: Directory is free to search, with 1,000 free credits a month. AyHire is $99 a month for one person, or Team at $79 a person a month, with a 14-day free trial and no card.

Our comparison of [GitHub recruiting tools](/blog/github-recruiting-tools) covers the other tools that read engineers' code.

## Which one should you use?

- **Firecrawl Directory** fits when the role isn't mostly code, when city or past employer is the main filter, or when you already have names and need work emails for very little.
- **AyHire** fits when the hard part is deciding which engineers are worth a call, and you want to see the code behind that decision before you write to anyone.
- **Both** fit in one chat, because they do different steps. Directory reaches people through their job history; AyHire checks what they built.

With both connected, one prompt can hand over from one to the other:

```
Use Firecrawl to find backend engineers who worked at companies like ours.
For each one with a public GitHub, use AyHire to judge their work against
my storage engineer role, and tell me who to talk to first.
```

One caveat holds for every tool that reads public work, AyHire included. Most engineers do their best work in private company repos. When someone's public work is thin, AyHire says a criterion is unknown instead of marking them down, and turns it into a question to ask. For those people, a profile may be the best signal you have.

## Where AyHire fits

Describe the role in a sentence, such as "a senior Go engineer who has built distributed storage". AyHire plans the GitHub searches and shows them before they run, finds the people who wrote most of the matching projects, and judges them against [the role's criteria](/guides/writing-criteria) from their commits, pull requests and READMEs, with a link behind every judgement. Then it tells you [who to talk to first](/blog/how-ayhire-decides), why, and what to check, and who is open to work.

It never contacts anyone for you, it never reads private repos, and it never uses where someone lives to score them.

AyHire is $99 a month for one person, or Team at $79 a person a month, with a 14-day free trial and no card. [Start your free trial](/pricing). To use it in the same Claude or ChatGPT chat as Firecrawl, [connect AyHire to your agent](/guide).

## Questions people ask

### What is Firecrawl Directory?

Firecrawl Directory is a recruiting data collection in Firecrawl's Alexandria library, launched on 28 October 2026. It gives ChatGPT and Claude access to more than 240 million professional profiles with job histories and work contact details, through the Firecrawl plugin. Its data comes from providers including Apollo, FullEnrich and Data Legion.

### Is Firecrawl Directory free?

Searching is free. Pulling detailed information and contact details costs credits, and every Firecrawl account gets 1,000 free credits a month without a card. At Firecrawl's 18 October prices, that is about 50 work emails or 30 full person matches a month.

### Does Firecrawl Directory use GitHub?

The launch post doesn't list GitHub as a source; it names Apollo, FullEnrich and Data Legion. It suggests a separate Firecrawl web search for someone's open-source contributions once you have found them. Nothing in the post judges that work against your role.

### Can I use Firecrawl and AyHire in the same chat?

Yes. Both connect to Claude and ChatGPT, so one conversation can use Firecrawl to find names and AyHire to read their public work. AyHire also works from coding agents such as Claude Code, Codex and Cursor.

### Is Firecrawl Directory good for hiring engineers?

It is a cheap way to find names and work emails, especially when location or past employer matters most. It shows where someone worked, not what they built, and Firecrawl says it isn't meant for evaluating their work. For that step, read their public work yourself or use a tool that reads it for you.

*Sources, all opened on 28 October 2026: [Introducing Directory](https://www.firecrawl.dev/blog/introducing-directory), [Firecrawl's launch on X](https://x.com/firecrawl/status/2115470932953219157), [How to Use Claude for People Search with Live Contact Data](https://www.firecrawl.dev/blog/claude-for-people-search) (18 October 2026), [Firecrawl pricing](https://www.firecrawl.dev/pricing). AyHire's search ran the same day; stars and commit counts are GitHub's.*