YC Jobs Scraper: Y Combinator Startup Jobs, Salary & Equity

Export jobs at Y Combinator startups from their public YC job pages, with salary and equity ranges parsed into numbers, visa and experience requirements, the company's YC batch, team size and industry, and the full job description.

No login, no browser, no personal data.

What you get

• Numbers, not text. $140K - $210K becomes salaryMin: 140000, salaryMax: 210000, salaryCurrency: "USD", plus a yearly equivalent for monthly pay. 0.25% - 2.00% becomes equityMinPercent: 0.25, equityMaxPercent: 2. USD, EUR, GBP, INR and CAD ranges are recognised.
• Company context in every row. YC batch (e.g. W24), one-liner, website, team size, year founded, HQ, industries, LinkedIn page and status.
• Full descriptions as Markdown and plain text, plus the interview process when the company lists one.
• Filters: roles, keywords, locations, remote only, visa (will sponsor / not required / US only), job type, YC batches, posted within, minimum yearly salary and currency, exclude words in the title.
• Only new jobs. Turn on Only new jobs since last run and schedule it daily or weekly for job alerts.
• Reports in the run's key-value store: COMPANIES (every company read, with status and job counts) and HEALTH (companies discovered, failures, empty-field ratios).

Three ways to run it

You want	Set
Jobs at specific YC companies	Companies: names or links, e.g. Stripe, https://www.ycombinator.com/companies/brex
A quick sweep of who is hiring	Leave Companies empty, Discovery: Fast. Reads YC's job listing pages: about 300 hiring companies and ~2,400 jobs
Every YC company	Discovery: Full. Also checks each company in YC's company sitemap (6,000+). Most complete, but slow

Filters narrow the output; Roles also narrows which listing pages Fast discovery reads.

How long a run takes

Pages are read one at a time with a pause in between, so run time grows with the number of pages:

Run	Typical time
A few companies, or Fast with a low Max jobs	under a minute
Fast, every job, descriptions off	about 7 minutes
Fast, every job, descriptions on	30 to 45 minutes
Full, descriptions off	1.5 to 2 hours
Full, descriptions on	2.5 to 3 hours

Descriptions need one extra page per job, so they are the main cost in time (not in price). For a full sweep, run it once with Include descriptions off, then run again with descriptions on for just the companies you care about. The default timeout is 4 hours; long runs save their Only new jobs memory as they go, so a stopped run never re-charges you for jobs you already received.

Output example

{
  "id": "yc:94746",
  "title": "Applied AI Engineer",
  "company": "Infer",
  "companyBatch": "S21",
  "companyOneLiner": "Operating system for insurance agencies",
  "companyTeamSize": 9,
  "role": "Engineering",
  "roleSpecificType": "Machine learning",
  "employmentType": "Full-time",
  "location": "Bengaluru, KA, IN / Bengaluru, Karnataka, IN",
  "isRemote": false,
  "minExperience": "3+ years",
  "visa": "US citizenship/visa not required",
  "salaryText": "₹2M - ₹5M INR",
  "salaryMin": 2000000, "salaryMax": 5000000, "salaryCurrency": "INR", "salaryPeriod": "annual",
  "equityMinPercent": 0.01, "equityMaxPercent": 0.2,
  "postedAgo": "5 months", "postedDaysAgo": 150,
  "descriptionText": "About us\n\nInfer is building the operating system for insurance agencies...",
  "url": "https://www.ycombinator.com/companies/infer/jobs/1DhbYdF-applied-ai-engineer"
}

Good to know

• Posting dates are approximate. YC only shows relative ages ("24 days", "about 2 months"), so postedDaysAgo and the Posted within filter are approximate.
• Unclear salaries are not guessed. A range without a unit such as €75 - €95 keeps its numbers but gets no yearly value, so salary filters skip it. The original text is always in salaryText.
• Applying. url and applyUrl point to the job's page on ycombinator.com; YC asks candidates to sign in there to apply.
• Not every YC company posts on YC. Some hire only through their own careers page. Those companies show jobs: 0 in the COMPANIES report.
• Personal data. Founder and hiring-manager names are never collected. Rows contain job and company data only.
• Polite by design. Pages are read one at a time with a pause between requests, and only pages YC's robots.txt allows. YC's own company search is not used.

Pricing

Pay per job returned plus a small start fee per run. Descriptions are included at the same price.

Feedback

Missing a field or a filter? Open an issue on this Actor's page. We read every one and reply.

Not affiliated with or endorsed by Y Combinator.