Y Combinator Companies Directory + Startup Jobs Scraper

The complete Y Combinator startup directory (≈6,000 companies since 2005) as a dataset, with the same search and filters as ycombinator.com/companies — batch, industry, status, region, tags, hiring, top companies, nonprofits — plus, on request, founders and social links per company and every open job at companies that are hiring (from the companies' YC jobs pages, the same listings as Work at a Startup).

It reads the directory's own public search index with the public search-only key that the YC website embeds for visitors (fetched fresh each run, never hard-coded) and the public company/jobs pages. No login, no API key, no browser, no proxies.

Typical uses: investor and sales prospecting (who raised what batch, where, hiring or not), competitor and market maps by industry/tag, job hunting at YC startups (salary + equity ranges are shown), founder outreach lists, research datasets, LLM enrichment.

What you get

One row per company (type: "company"):

field	description
id, name, slug, url, website	YC id, name, directory URL, company website
one_liner, long_description	Tagline and full description
batch, status, stage	e.g. Summer 2025, Active / Acquired / Public / Inactive, Early / Growth
industry, subindustry, industries, tags	e.g. B2B, B2B -> Sales, ["AI", "SaaS"]
regions, all_locations	e.g. ["United States of America", "America / Canada"], San Francisco, CA, USA
team_size, launched_at, is_hiring, top_company, nonprofit, former_names, logo_url

With Include company details each row also gets: founders[] {name, title, bio, linkedin_url, twitter_url}, linkedin_url, twitter_url, github_url, crunchbase_url, facebook_url, location, city, country, year_founded, primary_partner, job_count.

With Include open jobs you get one type: "job" row per open position at each hiring company: title, url, apply_url, company, company_slug, company_url, company_website, company_batch, company_one_liner, company_team_size, location, remote, job_type, role, role_type, salary_range (e.g. $140K - $175K), equity_range (e.g. 0.25% - 1.20%), min_experience, visa, skills, ask_us.

If the directory cannot be read (YC changed the page or search is unavailable) the run stores one type: "error" row — free.

Export as JSON, CSV or Excel, or push rows to Google Sheets, Make, Zapier, webhooks or your code via the Apify API.

Input

field	default	meaning
query	—	free-text search (name, tagline, description, website)
batches	all	e.g. ["Summer 2025", "Winter 2025"]
industries	all	industry or sub-industry names as on the site
statuses	all	Active, Acquired, Public, Inactive
regions	all	e.g. ["Europe", "United Kingdom"]
tags	all	e.g. ["AI", "Developer Tools"]
isHiring / topCompaniesOnly / nonprofitOnly	false	flag filters
includeDetails	false	founders, socials, year founded (1 request per company)
includeJobs	false	open jobs for hiring companies (1 request per hiring company)
maxItems	1000	max companies (job rows are extra)

Example — hiring AI companies from the two latest batches, with founders and jobs:

{
  "batches": ["Summer 2025", "Spring 2025"],
  "tags": ["AI"],
  "isHiring": true,
  "includeDetails": true,
  "includeJobs": true,
  "maxItems": 500
}

Whole directory: leave all filters empty and set maxItems to 10000 — the actor walks batch by batch to get past the search index's 1,000-hit page limit (≈6,000 rows, a couple of minutes).

Pricing

Pay per event: $0.001 per stored row — a company row or a job row ($1 per 1,000). Only rows actually written to the dataset are charged; error rows are free. Apify platform usage is billed separately by Apify (a full directory run is a few minutes of compute).

Limitations — please read

• Data is what YC publishes publicly: no funding amounts, valuations, revenue or contact e-mails. team_size and status are self-reported by companies and can be stale.
• Filter values must match the directory's own labels (Summer 2025, not S25; United States of America, not US). Unknown labels simply return 0 companies.
• Jobs come from each company's YC jobs page; job descriptions are not included (only the structured fields listed above and the link). Applying requires a Work at a Startup account, as on the site.
• Details and jobs cost one polite request per company (~2/s): 1,000 companies with both options ≈ 15–20 minutes.
• The public search key rotates; the actor reads it from the page on every run. If YC changes its site the run ends with a free error row instead of bad data.

Not affiliated with Y Combinator. Uses only the public website and the public search index it exposes to every visitor.