Lever Jobs Scraper: Live Board Data
Pricing
from $3.00 / 1,000 results
Lever Jobs Scraper: Live Board Data
Live job openings from any Lever-powered career board (jobs.lever.co): keyword search across companies, location, department and remote filters, full descriptions, and direct job links. Streams structured rows to your dataset in real time.
Pricing
from $3.00 / 1,000 results
Rating
0.0
(0)
Developer
Emmanuel
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
2 days ago
Last modified
Categories
Share
Lever Boards Real-Time Data
Free trial: runs on the Apify free plan return a 2-result sample per run. Upgrade to a paid Apify plan for full, unlimited exports.
Live job openings from any company career board hosted on Lever (jobs.lever.co) — board listings, keyword search, location and department filters, remote-only mode, full descriptions, and direct job links. Every row streams to your dataset the moment it is collected, and can also be pushed to a webhook in real time.
Who this is for
- Recruiters & sourcers building talent pipelines from live openings at target companies
- Staffing agencies monitoring hiring activity across dozens of Lever-powered employers
- Job-board & career-site operators keeping listings fresh without manual research
- Market researchers tracking hiring trends by company, department, location, and workplace type
- Founders & HR teams benchmarking competitor hiring and role definitions
- AI agents & automation builders feeding structured, up-to-date postings into CRMs, alerts, and workflows
Features
| Mode | What you get |
|---|---|
| 💼 Board jobs (default on) | Keyword search across one or more company boards: title, company, location, all locations, department, team, commitment, workplace type, country, posting date, descriptions, job + apply links |
| 🎯 Enrich with full details (toggle) | Refreshes each row with the latest full description and posting metadata. Adds a little extra time per record; off = fast board cards |
| 🔗 Job links | Paste individual Lever job URLs to collect the same structured row per link |
| ⚙️ Output & delivery | Streaming dataset writes, JSON or Slack webhooks, global max-items cap |
| 🌐 Connection | Apify residential US proxy on by default for reliable high-volume runs |
Input reference
Board jobs
| Field | Type | Default | Notes |
|---|---|---|---|
enableBoardJobs | checkbox | true | Master switch for board collection |
companySlugs | string list | ["spotify"] | Board names from jobs.lever.co URLs (e.g. spotify for jobs.lever.co/spotify). Accepts full board URLs too |
jobKeywords | string list | ["engineer"] | Matched against title, department, team, location, and description. Empty = all openings |
jobLocation | text | "" | Partial match on location (e.g. New York, Remote, London) |
jobDepartment | text | "" | Partial match on department or team (e.g. Engineering, Design) |
remoteOnly | checkbox | false | Keep only jobs marked remote |
jobDatePosted | select | any | any / month / week / day |
jobsPerCompany | number | 10 | Cap per company/keyword combination (1–200) |
searchFetchFullDetails | checkbox | false | Enrich every row in place (detailsFetched: true). Same row — never duplicated or dropped |
Job links
| Field | Type | Default | Notes |
|---|---|---|---|
enableScrapeByUrl | checkbox | false | Master switch for URL mode |
jobUrls | string list | [] | jobs.lever.co/COMPANY/ID links (US and EU boards supported) |
Output & delivery
| Field | Type | Default | Notes |
|---|---|---|---|
webhookUrl | text | "" | Optional HTTPS URL; each record is always saved to the dataset first |
webhookFormat | select | json | json = full record, slack = compact message |
Connection: standard Apify proxy object (Residential US by default).
Example — backend engineers in New York at two companies:
{"enableBoardJobs": true,"companySlugs": ["spotify", "ro"],"jobKeywords": ["backend engineer"],"jobLocation": "New York","remoteOnly": false,"jobsPerCompany": 20,"searchFetchFullDetails": true}
Output fields
One row per posting: featureType, jobId, companySlug, title, company, companyUrl, location, allLocations, department, team, commitment, workplaceType (onsite / remote / hybrid / flexible), country, jobUrl, applyUrl, descriptionText, descriptionHtml, postedAt, detailsFetched, searchQuery, searchCompany, sourceUrl, position, scrapedAt.
The run summary (counts, paywall status, spending-limit state) is stored in the key-value store under OUTPUT.
Webhooks
Set webhookUrl to receive one POST per record in addition to the dataset write. Use json for pipelines (CRM, Zapier, Make, Sheets) or slack for an incoming Slack webhook. Delivery is fire-and-forget — a failed webhook never stops the run.
MCP / AI-agent usage
Pair this Actor with the Apify MCP server to query live postings conversationally. Example prompt for your agent:
"Using the Lever Boards actor, find remote designer openings at Spotify and summarize the teams hiring."
FAQ
- Which companies are supported? Any employer whose careers page lives on
jobs.lever.co(US and EU boards). Paste the board name from the URL. - How fresh is the data? Every run reads the live boards, so rows reflect currently published openings.
- Do I need a proxy? Apify residential proxy is on by default and recommended for large runs. Small runs work without one.
- What does the free trial include? 2 results per run on the Apify free plan — enough to verify shape and fit. Paid plans are uncapped.
- How am I charged? Pay-per-event on the
resultevent (one per exported row), subject to your run's max-spend limit. When the limit is hit the run stops gracefully and keeps everything collected so far.