Lever Jobs Scraper: jobs.lever.co Postings Export

Collect current job postings from public Lever.co boards (jobs.lever.co, global and EU) to track hiring at target companies, find sales leads by open roles, or feed your own job board. Add a board slug, board URL, or one exact job URL. It uses the official Lever postings API: no API key, login or proxy. Get one row per unique job with a stable ID, description and published salary.

Start with targets: ["leverdemo"]. The new Input form has this small example filled in. Pricing is $0.002 per start + $0.45 per 1,000 useful job rows on the Free plan (down to $0.315 on Gold) at the default 512 MB memory. Check RUN_SUMMARY when a board is empty, a filter removes jobs, or a source request fails.

Table of contents

• What it does
• How it works
• Quick start
• Use cases
• Input
• Output
• Pricing
• Limits
• API, export and integrations
• Troubleshooting
• FAQ
• Related Actors
• Support

What it does

Read public published postings from known Lever boards. Global and EU boards are supported. A bare slug uses the global service. An exact posting URL reads that job. The Actor uses the public Lever Postings API with direct HTTPS requests. No Lever account or API key is needed.

Get titles, locations, departments, teams, commitments, workplace types and public links. Keep HTML and plain descriptions, list sections and published salary fields when includeContent is true. The Actor gives each job a stable ID and records outcomes for all input targets.

Good fit

• A snapshot of open jobs on known company Lever boards.
• Recruiting market research by team, location, commitment or workplace type.
• A job feed built from exported rows with stable IDs.
• A scheduled job alert: with onlyNewSinceLastRun each run returns only jobs that earlier runs have not returned.

This product does not discover companies, search all Lever jobs, read private/internal or closed postings, submit applications, or report closed or changed jobs. Other ATS platforms need a separate Actor.

👉 Try it now: open the input form

How it works

1. You give public Lever board slugs, board URLs or exact job URLs. Each one is checked and normalized; duplicates are read once.
2. The Actor reads the published jobs of each board from Lever's public postings service, global or EU, without a login, browser or proxy. Your filters and row caps are applied.
3. It saves one Dataset row per unique job with a stable ID, and a RUN_SUMMARY report with a status for every target: jobs found, empty board, filtered out, not found or a technical error.

👉 Try it now: open the input form

Quick start

1. Open the Input tab. Keep the filled targets example, leverdemo, or paste your public Lever board URL such as https://jobs.lever.co/leverdemo (box 1 in the picture). Click + Add for more companies in one run.

2. Keep Include descriptions and lists on for the full job text, lists and salary (box 2). Turn it off when titles, locations and links are enough.

3. Keep Maximum total jobs at the runtime default 5000 or set 100 for a first test (box 3). Filters are optional, see Input.

4. Optional: turn on Only new jobs since the last run for a scheduled job alert (box 4).

5. Click Start.

6. Open the Output tab (box 1 below). Each row is one job; check title, location, department and createdAt (box 2). Open Storage → Key-value store → RUN_SUMMARY for each target's status and counts.

   

Expected result: one Dataset row per published job with a stable id; the demo board had 10 published jobs when checked on 27 October 2026. Live jobs may change. API calls must supply targets; an empty API input does not select the demo.

👉 Try it now: open the input form

Use cases

Task	Input	Useful fields	Next step
Monitor hiring at target companies	Their Lever board slugs in targets, onlyNewSinceLastRun: true, run on a schedule	sourceJobId, title, team, createdAt, observedAt	Each run gives only new jobs; for closed jobs, compare id values of two full runs without onlyNewSinceLastRun
Find sales leads from open roles	Boards of prospect companies, filters.teams such as Sales or Engineering	title, department, team, locations, hostedUrl	Push rows to your CRM or sheet
Feed a job board or job alert	includeContent: true, filters.locationIncludes, filters.descriptionIncludes or filters.postedWithinDays	descriptionText, salaryRange, workplaceType, applyUrl	Publish or send the rows from your own system
Recruiting market research	Many boards, filters by team, location or commitment	team, department, commitment, country, salaryRange	Group rows by team, place or pay
Recheck a known posting	Exact job URL in targets	Stable id, observedAt, source fields	See whether the job is still published

👉 Try it now: open the input form

Input

targets is required. It must have 1–50 strings. Duplicate normalized targets are fetched once. Invalid targets get INVALID_INPUT entries and are not fetched. A missing/empty target array or invalid option shape fails validation.

Fields

Field	Type	Required	Runtime default / UI prefill	Allowed values and effect
targets	array	yes	No runtime default. UI prefill and schema default: ["leverdemo"].	Use a board slug such as leverdemo, a public Lever board URL, or one exact public job URL. Bare slugs use the global Lever service.
includeContent	boolean	no	Runtime default: true. No UI prefill.	Keep the public HTML/plain job content and list sections in each result.
filters	object	no	No filters. No UI prefill.	All filled filter groups must match. Values within one group use OR logic.
maxJobsPerTarget	integer	no	Runtime default: 0. No UI prefill.	Stop each board after this many useful results. Use 0 for no per-board cap. Range 0–5000.
maxTotalJobs	integer	no	Runtime default: 5000. No UI prefill.	Maximum number of unique job rows saved in one run. Range 1–5000.
pageSize	integer	no	Runtime default: 0. No UI prefill.	Use the complete board response when 0. A positive value uses Lever skip/limit pages. Range 0–1000.
continueOnError	boolean	no	Runtime default: true. No UI prefill.	Process independent targets after a technical failure and report a partial run.
onlyNewSinceLastRun	boolean	no	Runtime default: false. No UI prefill.	Return only jobs that an earlier run with the same targets and filters has not returned. See Only new jobs.

Nested filters

All filled groups must match (AND). Values within one group match any value (OR). Text matches ignore case and trim outer spaces. Empty arrays do not filter. Filters run after fetching source rows; a small result cap does not cap the source response size.

Field	Type / limit	Default	Match rule
filters.titleIncludes	array; up to 20 values	No restriction	Keep jobs whose title contains at least one of these values. Case-insensitive substring.
filters.titleExcludes	array; up to 20 values	No restriction	Skip jobs whose title contains any of these values, e.g. intern. Case-insensitive substring.
filters.locationIncludes	array; up to 20 values	No restriction	Keep jobs whose source location contains at least one of these values. Case-insensitive substring. Checks all locations.
filters.locationExcludes	array; up to 20 values	No restriction	Skip jobs with any location that contains one of these values. Case-insensitive substring.
filters.descriptionIncludes	array; up to 20 values	No restriction	Keep jobs whose posting text (opening, description, lists, additional) contains at least one value. Case-insensitive substring. Works with includeContent: false.
filters.descriptionExcludes	array; up to 20 values	No restriction	Skip jobs whose posting text contains any of these values. Case-insensitive substring.
filters.departments	array; up to 20 values	No restriction	Keep jobs whose source department equals one of these values. Case-insensitive exact value.
filters.teams	array; up to 20 values	No restriction	Keep jobs whose source team equals one of these values. Case-insensitive exact value.
filters.commitments	array; up to 20 values	No restriction	Keep jobs whose source commitment equals one of these values. Case-insensitive exact value.
filters.workplaceTypes	array; up to 4 values	No restriction	Keep jobs with one of these source workplace types.
filters.countries	array; up to 20 values	No restriction	Keep jobs whose source country is one of these ISO-2 values. Case-insensitive exact value.
filters.createdAfter	string; UTC ISO date-time, e.g. 2026-10-01T00:00:00Z	No restriction	Only jobs created on or after this ISO date-time. Jobs with no source creation time do not match.
filters.postedWithinDays	integer; 0–365	No restriction (0)	Only jobs created in the last N days, counted from the run time. Jobs with no source creation time do not match.

Filter strings have a maximum length of 200 characters; country values use two letters. workplaceTypes accepts remote, hybrid, on-site, unspecified.

Only new jobs

With onlyNewSinceLastRun: true the Actor remembers the job IDs it has handled and skips them in later runs. The memory is a named key-value store in your Apify account, lever-jobs-scraper-state, with one record per set of targets and filters. The first run returns all matching jobs as a baseline. A different target list or different filters start a new baseline. Jobs cut off by maxJobsPerTarget, maxTotalJobs or a charge limit stay new for the next run, and so do jobs that failed to save. Up to 50,000 IDs are kept per record; the oldest are dropped first. Delete the record to start over.

RUN_SUMMARY shows previouslySeenJobs, stateBaseline (true on the first run) and stateSaved. A board with no new jobs gets FILTERED_EMPTY. Skipped known jobs are billed like filtered jobs (filtered-job, $0.01 per 1,000), and a board with no new jobs gets one no-result-target.

Setting interactions

Settings	Result
maxJobsPerTarget: 0 + maxTotalJobs: 5000	No per-board cap; stop at 5,000 unique useful rows across the run
Positive per-target and total caps	Apply both caps after filtering and dedupe; a per-target cap still allows later boards
pageSize: 0	Read one complete board response, subject to the 25 MiB response limit
Positive pageSize	Use Lever skip/limit pages; stop at a short/empty page, cap or repeated-page error
includeContent: false	Omit content and salary fields; lists is []; identity/category/link fields remain
continueOnError: true	Continue independent targets after a source failure and show a partial summary
onlyNewSinceLastRun: true	Jobs already returned by an earlier run with the same targets and filters are skipped and billed as filtered-job

Examples

Small demo, also used for the output example below:

{"targets": ["leverdemo"]}

Two boards with a title filter and a per-board cap:

{
  "targets": ["leverdemo", "https://jobs.lever.co/palantir"],
  "filters": {"titleIncludes": ["engineer"]},
  "maxJobsPerTarget": 20,
  "maxTotalJobs": 40,
  "includeContent": true
}

This saves up to 20 matching jobs from each board and 40 in total.

Job alert for remote engineering jobs that mention Kubernetes, without interns, only new since the last run:

{
  "targets": ["leverdemo", "https://jobs.lever.co/palantir"],
  "filters": {
    "titleIncludes": ["engineer"],
    "titleExcludes": ["intern"],
    "descriptionIncludes": ["kubernetes"],
    "workplaceTypes": ["remote"],
    "postedWithinDays": 30
  },
  "onlyNewSinceLastRun": true
}

A filter may return no matches. EU input uses https://jobs.eu.lever.co/<board>. Public api.lever.co/v0/postings/<board> and EU API URLs are also accepted. HTTPS is required. Arbitrary hosts, userinfo, custom ports, fragments and traversal paths are rejected. Tracking query parameters are discarded.

👉 Try it now: open the input form

Output

Dataset fields

Each row is one unique published job. The stable ID is lever:{region}:{boardSlug}:{encodedSourceJobId}. The same title can occur on distinct jobs; IDs control dedupe. observedAt is this Actor's observation time. createdAt comes from the source, converted to an ISO timestamp; it is not an update time.

Field	Meaning	Type	Can be empty?	Example / enum	Source / notes
id	Stable ID: lever:{region}:{board}:{postingId}.	string	no	lever:global:leverdemo:ad208490-…	Set by the Actor
source	The public source name.	string	no	lever	Set by the Actor
idRuleVersion	Version of the stable identifier rule.	string	no	v1	Set by the Actor
region	Global or EU public Lever instance.	string	no	global, eu	Set by the Actor
boardSlug	Normalized public Lever site slug.	string	no	leverdemo	Set by the Actor
boardUrl	Canonical public hosted board URL.	string	no	https://jobs.lever.co/leverdemo	Set by the Actor
sourceJobId	Upstream Lever posting ID.	string	no	ad208490-4052-4f91-a57e-433f2d1e484b	Lever source, not guessed
title	Published job title.	string	no	Customer Success Manager	Lever source, not guessed
location	Primary source location, when present.	string	yes (null)	Atlanta, Georgia	Lever source, not guessed
locations	Source location values.	array	can be []	["Atlanta, Georgia"]	Lever source, not guessed
department	Source department, when present.	string	yes (null)	Customer Success	Lever source, not guessed
team	Source team, when present.	string	yes (null)	Customer Success Team	Lever source, not guessed
commitment	Source commitment, when present.	string	yes (null)	Regular Full Time (Salary)	Lever source, not guessed
country	Source ISO-2 country, when present.	string	yes (null)	US	Lever source, not guessed
workplaceType	Source workplace type or unspecified.	string	no	remote, hybrid, on-site, unspecified	Lever source, not guessed
openingHtml	Source opening HTML when content is included.	string	yes; omitted with includeContent: false	text or HTML from the posting	Lever source, not guessed
openingText	Plain opening text when content is included.	string	yes; omitted with includeContent: false	text or HTML from the posting	Lever plain field, or converted from HTML
descriptionHtml	Source combined description HTML.	string	yes; omitted with includeContent: false	text or HTML from the posting	Lever source, not guessed
descriptionText	Plain combined description.	string	yes; omitted with includeContent: false	text or HTML from the posting	Lever plain field, or converted from HTML
descriptionBodyHtml	Source body HTML.	string	yes; omitted with includeContent: false	text or HTML from the posting	Lever source, not guessed
descriptionBodyText	Plain body text.	string	yes; omitted with includeContent: false	text or HTML from the posting	Lever plain field, or converted from HTML
lists	Source list sections with title, HTML and text.	array	can be []; [] with includeContent: false	[{"title": "Requirements", …}]	Lever source, not guessed
additionalHtml	Source additional HTML.	string	yes; omitted with includeContent: false	text or HTML from the posting	Lever source, not guessed
additionalText	Plain additional text.	string	yes; omitted with includeContent: false	text or HTML from the posting	Lever plain field, or converted from HTML
salaryRange	Source salary range when provided.	object	yes; omitted with includeContent: false	min, max (number or null), currency, interval (string or null)	Lever source, not guessed
salaryDescriptionHtml	Source salary note HTML.	string	yes; omitted with includeContent: false	text or HTML from the posting	Lever source, not guessed
salaryDescriptionText	Plain salary note.	string	yes; omitted with includeContent: false	text or HTML from the posting	Lever plain field, or converted from HTML
hostedUrl	Canonical public job page.	string	no	https://jobs.lever.co/leverdemo/ad208490-…	Lever source, not guessed
applyUrl	Source application URL; never submitted by the Actor.	string	yes (null)	https://jobs.lever.co/leverdemo/ad208490-…/apply	Lever source, not guessed
createdAt	Source creation time in ISO format.	string	yes (null)	2026-09-09T11:01:39.000Z	Lever source, not guessed
observedAt	Time this Actor observed the row.	string	no	2026-10-28T05:00:00.000Z	Set by the Actor

lists[] has title (section label), html (section body), and text (plain body), all strings. salaryRange is null or an object with min/max (number or null), currency/interval (string or null). No salary or location is guessed. workplaceType falls back to unspecified.

Missing source scalar values become null. Arrays can be empty. With includeContent: false, HTML/text and salary fields are omitted rather than set to null. HTML is source content; sanitize it before displaying it in your own web app. Plain text uses source plain fields when supplied, or HTML conversion. List body text is converted from list HTML. Some uncommon named HTML entities can remain encoded.

Example row

This trimmed row is from the leverdemo input above. Full descriptions and the remaining fields are in the Dataset.

{
  "id": "lever:global:leverdemo:ad208490-4052-4f91-a57e-433f2d1e484b",
  "sourceJobId": "ad208490-4052-4f91-a57e-433f2d1e484b",
  "boardSlug": "leverdemo",
  "title": "Customer Success Manager",
  "location": "Atlanta, Georgia",
  "locations": [
    "Atlanta, Georgia"
  ],
  "department": "Customer Success",
  "team": "Customer Success Team",
  "commitment": "Regular Full Time (Salary)",
  "workplaceType": "hybrid",
  "hostedUrl": "https://jobs.lever.co/leverdemo/ad208490-4052-4f91-a57e-433f2d1e484b",
  "applyUrl": "https://jobs.lever.co/leverdemo/ad208490-4052-4f91-a57e-433f2d1e484b/apply"
}

Reports and outcomes

The Output tab links to dataset and runSummary. They are links, not a second copy of the rows. The Actor writes RUN_SUMMARY in the default key-value store; it does not write a separate OUTPUT record.

RUN_SUMMARY includes schemaVersion, status, received/valid/duplicate/processed/unprocessed target counts, results, duplicateJobs, emptyBoards, filteredEmpty, notFound, technicalFailures, retries, rate-limit counts, invalid/filtered source row counts, previouslySeenJobs, stateBaseline and stateSaved (only with onlyNewSinceLastRun), limitReached, chargeLimitStop, charge events/mode, transport, timings, requests, bytes, failed target references, and targets[].

Each target entry has inputIndex, canonicalTarget, mode, region, status, source/emitted row counts, retries and errorCode; bytes/requests and duplicate references are present when relevant.

Outcome	Dataset / report	What to do
OK	Useful job rows	Export or use rows
EMPTY_BOARD	No jobs returned by a valid source board	Check later if needed
FILTERED_EMPTY	Source had jobs, filters matched none	Review filters
NOT_FOUND	Board or exact posting returned HTTP 404	Check the URL; posting may be closed
INVALID_INPUT, DUPLICATE_TARGET	No result row for that target	Correct invalid input; duplicates need no retry
LIMIT_REACHED	Saved rows retained; more jobs may exist	Review per-target/total/platform charge caps
Technical error	TIMEOUT, RATE_LIMITED, UPSTREAM_5XX, UPSTREAM_4XX, NETWORK_ERROR, BAD_CONTENT_TYPE, MALFORMED_JSON, SCHEMA_DRIFT, RESPONSE_TOO_LARGE, REDIRECT_ERROR, STORAGE_ERROR, ABORTED or UNPROCESSED	Read RUN_SUMMARY; fix input or retry later

Run-summary status is SUCCEEDED, PARTIAL, FAILED or LIMIT_REACHED. Apify run status and summary status are separate: a partial run can finish on the platform and still have failed targets. A full technical source failure fails the platform run. Empty/missing targets do not prove that a company has no jobs elsewhere.

👉 Try it now: open the input form

Pricing

Pay per event. Apify charges the standard apify-actor-start event automatically: $0.002 per start, one event per GB of run memory (minimum one). At the default 512 MB each run has one start event; 2 GB means two.

Event	Free plan	Bronze	Silver	Gold
Unique job row saved (unique-job-result)	$0.45 / 1,000	$0.405 / 1,000	$0.36 / 1,000	$0.315 / 1,000
Job read but skipped by your filters (filtered-job)	$0.01 / 1,000	same	same	same
Board or job URL with no paid rows (no-result-target)	$0.0003 each	same	same	same

no-result-target is charged once for a board or exact job URL that is not found, empty, or fully filtered. Duplicate rows, duplicate targets, invalid input and technical failures (timeouts, rate limits, Lever errors) are never charged. A partial run pays the start plus accepted useful rows and the small read events above.

Run at 512 MB (Free plan)	Actor event charge
One missing board, 0 rows	$0.0023
10 useful jobs	$0.0065
100 useful jobs	$0.047
1,000 useful jobs	$0.452
5,000 useful jobs (current maximum per run)	$2.252
A 4,000-job board filtered down to 0 rows	$0.0423
A 4,000-job board filtered down to 100 rows	$0.086

The filtered-job fee applies to every job your filters skip, also when other jobs match. On a very large board with a narrow filter it can be a noticeable part of the bill.

These are event fees; the standard PPE tariff has no separate customer platform-usage surcharge. Check the run billing and your plan.

maxTotalJobs caps rows, not a bill in dollars. Use Apify's maximum total charge option (maxTotalChargeUsd in API calls) to limit total Actor event charges, including the start. At 512 MB, allow at least $0.00245 for one start and one useful job. The run stops cleanly when the limit is reached. This option does not limit all platform costs.

RUN_SUMMARY.chargeEvents counts job-result events; filteredJobEvents and noResultTargetEvents count the two small read events. The automatic start appears in the Apify run billing. The summary is not a customer invoice.

👉 Try it now: open the input form

Limits

• Up to 50 targets and 5,000 saved jobs per run.
• Requests are paced one at a time with a 1-second host gap, up to 3 attempts, 20 seconds per request including body reading, up to 2 safe redirects, 25 MiB per response, and 1,000 pages per board. Large boards or repeated retries may exceed your platform timeout. Completeness means the public source response observed for your supplied boards, subject to these limits. Jobs may change during pagination.
• Only public published postings of boards you supply: no company discovery, global Lever search, private/internal or closed postings, applications, or closed/changed-job reports.
• Salary, location and dates are only what the source publishes; nothing is guessed.

👉 Try it now: open the input form

API, export and integrations

• API: start a run with the Apify Actor run API and the same input JSON, then read the default Dataset.
• Export: Dataset export supports JSON and CSV, plus other Apify formats. Nested lists are best kept in JSON.
• Scheduling: you can use Apify scheduling with the same input. Each run has a start charge. Add onlyNewSinceLastRun: true to get only new jobs in each scheduled run; closed or changed jobs are not reported, so use the stable id in your system for those.

Use with AI agents (MCP)

• MCP server with only this Actor: https://mcp.apify.com?tools=cliqtomedia/lever-jobs-scraper.
• Smallest useful input: {"targets": ["leverdemo"]}.
• Key output fields: id (stable job ID), title, location, team, hostedUrl (public job page); RUN_SUMMARY.targets[].status tells why a board gave no rows.
• A small board such as the demo returns about 10 rows for about $0.0065 at 512 MB on the Free plan. Up to 50 targets and 5,000 rows per run; supply known board slugs, the Actor does not search Lever.

👉 Try it now: open the input form

Troubleshooting

• URL rejected: use a supported HTTPS Lever host and one board or posting path. Do not paste a company's general careers URL.
• No rows: open RUN_SUMMARY. Check EMPTY_BOARD, FILTERED_EMPTY, NOT_FOUND and technical errors separately.
• Fewer rows: check caps, chargeLimitStop, filters and per-target status. An early stop does not prove a complete board.
• Timeout, block or rate limit: retry later with fewer targets. No login, browser or proxy bypass is provided.
• Missing salary or dates: the source may not publish them. They are not inferred.

👉 Try it now: open the input form

FAQ

Do I need a Lever account?

No. Only public published postings are read. See Lever Postings API documentation.

Can I get every job on Lever?

You must supply known boards. There is no company discovery or cross-company global search.

Can I get EU jobs or one job URL?

Yes. Use an EU hosted board/job URL for EU mode. An exact posting URL returns that job if it remains public.

Does a repeat run return only new jobs?

Only if you turn on onlyNewSinceLastRun. Then the first run returns all matching jobs and later runs with the same targets and filters return only jobs not returned before. Without it, each run is a fresh snapshot.

Is no output free?

No. A zero-row run still has the $0.002 start charge at default 512 MB, plus $0.0003 for each board or job URL that gave no rows and $0.01 per 1,000 jobs skipped by your filters; there is no job-result event. Higher memory can increase the start count. Check your run's billing.

Can descriptions or locations be missing?

Yes. The source controls those fields. Check nulls, empty arrays and includeContent; do not treat missing fields as guessed values.

👉 Try it now: open the input form

Related Actors

• Greenhouse Jobs Scraper & Monitor — the same job rows from companies that use Greenhouse boards.
• Ashby Jobs Scraper — public Ashby boards, for companies on Ashby instead of Lever.
• Workable Jobs Monitor & Scraper — Workable career pages, with job changes between runs.
• SmartRecruiters Jobs Scraper — SmartRecruiters boards, often used by large employers.
• Remote Jobs Aggregator — remote jobs from RemoteOK, We Work Remotely and Himalayas when you do not have a company list yet.

👉 Try it now: open the input form

Support

Open an issue in this Actor's Apify Issues tab. Include the run ID, public input mode and the relevant RUN_SUMMARY target status. Do not share tokens, cookies, private URLs or sensitive payloads.

Lever identifies the source format. This product is not affiliated with or endorsed by Lever.

👉 Try it now: open the input form