Lever Job Scraper

Every open posting from any Lever job site, as clean JSON. Paste a jobs.lever.co URL or just the site name and get title, team, department, location, workplace type, commitment, structured salary range, posted date, the full description and the apply link for every posting.

Built for recruiters sourcing from technology and growth companies, sales teams that use hiring as a buying signal, job boards that need a direct feed, and AI agents that need job data in one shape.

Why a dedicated Lever scraper

Lever publishes a public postings API, so the data is reachable. This actor does the part that takes a day to get right:

• One request per site returns every posting with its full description and structured salary range when the company publishes one; no per-job calls.
• EU-hosted sites (jobs.eu.lever.co) are detected and read from the EU API automatically.
• Lever's categories (team, department, location, commitment, level, all locations) are flattened into the same fields every other FinData job scraper uses, so a Lever posting and a Workday posting look the same downstream.
• Workplace type (remote, hybrid, on-site) becomes a plain remote flag.
• Only-new mode, filters that run before billing, and a fail-soft run that finishes even when one site has been renamed or closed.

Finding a Lever site

Open the company's careers page and click any job. If the address bar shows lever.co, the first path segment is the site name. Any of these work as a source:

Shape	Example
Site URL	https://jobs.lever.co/palantir
A posting page	https://jobs.lever.co/zoox/3f1c2a9e-...
EU site	https://jobs.eu.lever.co/company
Just the site name	veeva

Companies that embed Lever on their own domain still have a site name; it is in the apply link (jobs.lever.co/{site}/{id}/apply).

Input

{
  "sources": ["palantir", "https://jobs.lever.co/zoox"],
  "keywords": ["software engineer", "machine learning"],
  "locations": ["Denver", "Remote"],
  "postedWithinDays": 7,
  "onlyNew": true
}

Field	Type	Default	What it does
preset	string	custom	Employer list: custom (default), sample. custom runs the sites you list and falls back to the sample boards when the list is empty
sources	string[]	[]	Lever site URLs or site names
keywords	string[]	[]	Keep postings whose title contains any of these (case-insensitive)
excludeKeywords	string[]	[]	Drop postings whose title contains any of these
locations	string[]	[]	Keep postings whose location text contains any of these
departments	string[]	[]	Keep postings whose department or team contains any of these
remoteOnly	boolean	false	Keep only postings Lever marks remote or whose location says remote
postedWithinDays	integer	0	Keep only postings created or updated in the last N days; 0 disables
includeDescriptionHtml	boolean	false	Add the original HTML description next to the plain text
onlyNew	boolean	false	Output only postings not seen on a previous run of this site
seenStoreName	string	lever-job-scraper-seen	Key-value store that remembers seen posting IDs; use one name per monitor
maxJobsPerSource	integer	0	Stop after N postings per site; 0 means all
maxConcurrency	integer	5	Sites pulled in parallel

Descriptions come with the list call, so there is no "fetch details" switch; every record has descriptionText.

Output

One record per posting.

{
  "vendor": "lever",
  "source": "https://jobs.lever.co/palantir",
  "company": "palantir",
  "jobId": "3f1c2a9e-5b7d-4c21-9e0a-8d2f6c1b4a77",
  "title": "Software Engineer, Backend",
  "department": "Engineering",
  "team": "Engineering",
  "location": "Denver, CO",
  "locations": ["Denver, CO", "Washington, D.C."],
  "country": "US",
  "remote": false,
  "employmentType": "Full-time",
  "postedAt": "2026-10-20T16:40:12.000Z",
  "updatedAt": "2026-11-02T11:03:55.000Z",
  "url": "https://jobs.lever.co/palantir/3f1c2a9e-5b7d-4c21-9e0a-8d2f6c1b4a77",
  "applyUrl": "https://jobs.lever.co/palantir/3f1c2a9e-5b7d-4c21-9e0a-8d2f6c1b4a77/apply",
  "descriptionText": "A World-Changing Company...",
  "salaryMin": 140000,
  "salaryMax": 200000,
  "salaryCurrency": "USD",
  "salaryPeriod": "year",
  "salaryText": "Salary range: $140,000 - $200,000",
  "extra": { "level": null, "workplaceType": "onsite" },
  "scrapedAt": "2026-11-05T01:47:38.912Z"
}

Field	Meaning
company	The Lever site name (Lever's public API does not carry a display name)
jobId	Lever posting UUID. Stable; use it as the key
department, team	Lever's department and team categories
location, locations	Lever's primary location, then every entry in allLocations
country	Lever's country code when set
remote	true for workplace type remote or a location that says remote; false for on-site or hybrid; null when Lever does not say
employmentType	Lever commitment (Full-time, Part-time, Intern, Contract)
postedAt, updatedAt	createdAt and updatedAt from Lever, ISO 8601
salaryMin, salaryMax, salaryCurrency, salaryPeriod, salaryText	Lever's structured salaryRange and salaryDescriptionPlain when the company publishes them; otherwise null
extra.level, extra.workplaceType	Lever's level category and raw workplace type

A RUN_SUMMARY record in the default key-value store reports sites processed, postings fetched, filtered, de-duplicated, skipped as already seen, and any site failures with the error text.

Pricing

Pay per event. You are charged per posting written to the dataset, plus a small actor-start event. Filtered-out and already-seen postings are not charged. One request per site and no proxies, so platform usage is a fraction of a cent per run.

Monitoring new postings

1. Set onlyNew to true and give seenStoreName a name for this monitor.
2. Add your filters.
3. Schedule the actor (daily, or hourly for sites you care about).
4. Connect the dataset to Google Sheets, Slack, Make, Zapier or a webhook from the Integrations tab.

The first run outputs everything on the sites and remembers it. Each later run outputs only what is new.

Limits and notes

• Public postings API only. No login, no proxies, no candidate or private data.
• Sites that are closed or renamed return an error for that site only; the run continues.
• Lever's public API does not expose a company display name; company is the site name.
• This actor reads Lever sites only. For Workday, Greenhouse, Ashby or SmartRecruiters, or for a mixed list where you do not know the ATS, use the FinData Career Site Job Scraper.

Related FinData job scrapers

Same engine, same output fields, so records from any of them can be combined.

• Workday Job Scraper
• Greenhouse Job Scraper
• Ashby Job Scraper
• Career Site Job Scraper (any company, ATS auto-detected)
• Workday, Greenhouse, Lever, Ashby, SmartRecruiters Job Scraper (all five in one actor)

For AI agents

Every field is optional: an empty input {} runs the sample boards and returns jobs, so a first call never fails. Pick "preset": "sample" to say so explicitly. Minimal call: { "sources": ["palantir"] }. Add "keywords" and "locations" arrays to narrow. Every record has the same fields and includes the description; jobId is the stable key.

Using it from code

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: process.env.APIFY_TOKEN });
const run = await client.actor('findata/lever-job-scraper').call({
    sources: ['palantir', 'zoox'],
    keywords: ['engineer'],
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();

Works the same through the Apify API, MCP server and the Python client.