Lever Jobs Scraper

What this Actor scrapes

Any employer running its careers page on Lever serves the entire board from one endpoint: https://api.lever.co/v0/postings/{token}. This Actor hits it directly — hand it one or more company tokens (the {token} in https://jobs.lever.co/{token}, e.g. palantir) and it pulls every currently-open posting for that employer in a single response, no pagination to chase, and normalizes it into one row schema. We've verified it end to end against real boards ranging from a zero-posting tenant to a 388-posting demo board, so the same code path holds whether a company has zero reqs open or hundreds.

What we handle for you

• 🔑 Keyless public API — no OAuth, no token, no per-company setup; the same JSON endpoint Lever's own hosted careers page renders from.
• 🏢 Multi-company in one run — pass many tokens; rows come back identically shaped, tagged with company.
• 🧭 Token or full URL, either works — paste a bare company token or the full https://jobs.lever.co/{token} link and we extract the right identifier from the URL path.
• 🎛️ Server-side filtering — narrow by team, department, location, or commitment before the payload is even sent, instead of pulling everything and filtering client-side.
• 🚫 Clean invalid-token detection — an unknown or unused token gets a plain 404, and we skip that company and keep going instead of choking on it or the run silently producing nothing.
• 🧩 Independent team + department fields — never collapsed into a single fallback like a shallower multi-ATS integration might do; both fields, and the full remote / on-site / hybrid workplace-type enum, come through untouched.
• 📋 Full requirement lists — the lists[] responsibilities/requirements blocks ship HTML-stripped to plain text, alongside the complete job description.
• 🔁 Retries with backoff + fingerprint rotation — curl-cffi presents a real browser handshake and retries transient 429/5xx responses with exponential backoff up to 5 attempts, so a large multi-company batch finishes instead of dying on the first hiccup.
• 🌐 Proxy-routed through Apify Proxy — every request runs through a rotating session, with a fresh exit IP requested the moment anything looks off — the same infrastructure we lean on harder the day a target tightens up.
• 🧊 Clean, typed rows — Pydantic-validated, ISO-8601 timestamps, stable posting IDs. Export JSON / CSV / Excel straight from the Apify Console.
• 💰 Pay only for postings that land — a company with an invalid token or zero open reqs never gets billed for rows it didn't produce.

Use cases

• Recruiting & talent intelligence — track what a Lever-hosted employer is hiring for, on which team, in which department, and under what workplace arrangement.
• Job-board aggregation — add Lever coverage alongside Greenhouse / Ashby / Workday / Teamtailor in one pipeline.
• Labor-market research — sample hiring demand across an industry or region by team, department, and commitment type.
• Hiring-intent signals for BD/SDR teams — a sudden spike in open reqs at a target account is a lead qualifier.
• HR-tech data pipelines — wire structured job rows, including full requirement text, into your CRM, dashboard, or n8n / Make workflow on a schedule.

Input

Paste this into the Apify Console, or send it as the run input over the API. Proxy settings are on by default; you rarely need to touch them.

{
  "companies": [
    "leverdemo",
    "plaid"
  ],
  "maxItems": 200,
  "proxyConfiguration": {
    "useApifyProxy": true
  }
}

Output

One row per result, schema-validated before it is written. Export JSON, CSV, Excel or XML from the run, or read it over the API.

posting_idcompanytitleteamdepartmentlocationworkplace_typecommitmenthosted_urlapply_urlcreated_atdescriptionlistsscraped_at

{
  "posting_id": "ac978161-6f46-4f6b-ad9e-a258e642751c",
  "company": "palantir",
  "title": "Administrative Business Partner",
  "team": "Administrative",
  "department": null,
  "location": "London, United Kingdom",
  "workplace_type": "hybrid",
  "commitment": "Full-time",
  "hosted_url": "https://jobs.lever.co/palantir/ac978161-6f46-4f6b-ad9e-a258e642751c",
  "apply_url": "https://jobs.lever.co/palantir/ac978161-6f46-4f6b-ad9e-a258e642751c/apply",
  "created_at": "2024-03-25T22:30:16.463Z",
  "description": "A World-Changing Company\n...",
  "lists": [
    {
      "text": "What We Require",
      "content": "At least three years...\nExcellent digital literacy..."
    }
  ],
  "scraped_at": "2026-07-29T15:00:00Z"
}

Pricing

You pay only for results that land. Cap any run with maxTotalChargeUsd. See pricing & billing for worked examples.

Limitations

• Public postings only — the openly published careers board, not authenticated internal/employee views.
• Company tokens, not company names — you supply the Lever token from the careers URL; a display name like "Palantir" may not match its actual token.
• No multi-location breakout — location returns the primary posting location; the full allLocations[] array for multi-location postings isn't row-exploded in v1.
• Salary is not modeled — some Lever tenants expose salary text, others don't; this Actor doesn't parse or emit it in v1 either way.
• Point-in-time snapshot — returns the board as it stands right now; schedule recurring runs to track changes.