Lever Job Delta Monitor: New Postings From Lever-Hosted Startups

This actor watches company slugs on Lever and returns a row for every posting it has not shown you before. Each row carries the team, the commitment (permanent, contract, intern) and Lever's own remote, hybrid or on-site flag. It calls Lever's public postings API, so there is no login and no browser involved.

Who it's for

A curator of a weekly remote-jobs newsletter. You follow about sixty startups that recruit through Lever. Your readers want this week's fresh remote roles, not the whole backlog again. Before this you opened sixty tabs on Friday; now you read one dataset.

An outplacement coach. Clients who were laid off give you a short list of dream employers. You schedule a run every morning and forward the new roles that match each client's team, such as design or data, so nobody has to refresh career pages while also job-hunting.

A bootcamp career-services lead. You maintain a list of companies that hire junior engineers and want to know the same day one of them opens an internship or an entry-level req.

What these three share is a standing list of employers and a need to see changes, not totals.

Why this one

• Lever's own classification comes through. team, commitment and workplace_type are separate columns, so a remote-only or contract-only view is a filter in your spreadsheet, not a text search.
• Billed on what you receive. The unit is one posting returned. A morning with no new roles creates no rows and no job charges.
• Your seen-list stays yours. The record of what you have already received sits in a named key-value store in your own Apify account.
• A cost ceiling that loses nothing. If a run stops at your maximum-cost limit, postings that were not delivered remain unseen and arrive on the next run.

What you get

Field	Meaning
job_id	Lever's posting id (a UUID). The baseline is keyed on it.
title	Posting title.
location	The primary location on the posting.
team	The team label Lever stores, for example "Experience".
commitment	The commitment label, for example "Permanent".
workplace_type	remote, hybrid or onsite as Lever reports it.
created_at	When the posting was created, converted to an ISO timestamp.
url	The hosted posting page, which includes the apply button.
lever_company_id	The slug the row came from.
first_seen_timestamp	When this monitor first recorded the posting.

One measurement to calibrate expectations: on 2026-10-08, Spotify's Lever board held 74 postings, 28 remote, 36 hybrid and 10 on-site.

What your first run costs, and how to start silently

A watchlist begins with a backlog, and the backlog is billed like any other batch of postings. Three ways to handle it:

• Accept it. Spotify's 74 postings come to about $0.07 on the first run. For ten small startups it is usually cents. This is the right choice when you want the full current picture before the deltas begin.
• Seed silently. Set skipFirstRun to true. The first run records every posting id for each slug and returns nothing, so the first rows you pay for are genuine arrivals. Pick this when your list contains large employers.
• Pre-filter. Add includeKeywords, such as "designer", before the first run. The filter is applied before anything is remembered, so only matching postings enter the seen-list and you are billed for those alone. Change the filter later and the postings it used to hide arrive as new on the next run, with no baseline reset required.

After the first run, each run compares the live board with the stored record for that slug and returns only the missing ids that pass your filter, up to maxResults; ids beyond the cap, or ones that could not be billed or saved, are not recorded and surface on the following run. The record lives in the key-value store lever-delta-store, or in the store named by deltaName, under the key baseline-<slug>. Deleting that key makes the slug start over.

Pricing

• Job posting returned: $1 per 1,000 job postings

Plus a $0.001 start fee per run. Each event above is billed independently, only when it actually returns data — misses (found:false) are never charged.

For a steady-state watchlist the bill is dominated by genuine arrivals. Twenty new roles in a week cost two cents.

How to use

1. In the Apify Console. Open the actor page and click Start — the lever_company_id field is already pre-filled with a working example. Results land in the run's dataset as soon as each item is found.
2. Via the API. Call it directly with a POST request — no Console needed once you have an API token:

   curl "https://api.apify.com/v2/acts/accountable_eel~lever-job-delta-monitor/run-sync-get-dataset-items?token=<YOUR_TOKEN>" \
     -X POST \
     -H "Content-Type: application/json" \
     -d '{"lever_company_id":["spotify"]}'
   

3. On a schedule. Save this actor as an Apify Task with the input you want, then add a Schedule (hourly, daily, weekly) so it runs on its own — no server of your own required.

The slug is the last part of the board address: for jobs.lever.co/spotify it is spotify. You can paste either form.

Input

{
  "lever_company_id": [
    "spotify"
  ]
}

Lever company slug or full URL (e.g. "spotify" or "https://jobs.lever.co/spotify"). Charged per job posting returned. Boards with no result are free. Accepted formats: spotify, https://jobs.lever.co/spotify.

Further settings: includeKeywords and excludeKeywords (matched case-insensitively against title, location and team), maxResults, testRun, onlyFound, deltaMode, deltaName and skipFirstRun. The field company_board_id is accepted as an alias for lever_company_id.

Sample output

query	found	status	job_id	title	location	team	commitment	workplace_type	created_at	url	lever_company_id	first_seen_timestamp	scrapedAt
spotify	true	OK			<location>								1970-01-01T00:00:00.000Z

A slug that Lever does not recognise returns one found: false row with the status NOT_FOUND, and any other API error comes back as ERROR. Neither is billed.

Use it from Clay, n8n, Make, or an AI agent

This actor runs synchronously over plain HTTP — call it directly from a script, a workflow tool, or an AI agent, no Apify Console needed once you have an API token.

curl "https://api.apify.com/v2/acts/accountable_eel~lever-job-delta-monitor/run-sync-get-dataset-items?token=<YOUR_TOKEN>" \
  -X POST \
  -H "Content-Type: application/json" \
  -d '{"lever_company_id":["spotify"]}'

n8n. Add an HTTP Request node: Method POST, URL https://api.apify.com/v2/acts/accountable_eel~lever-job-delta-monitor/run-sync-get-dataset-items?token=<YOUR_TOKEN>, Body Content Type JSON, JSON Body {"lever_company_id":["spotify"]} (swap in an expression from an earlier node for a real value).

Clay. Add an "HTTP API" column: Method POST, URL https://api.apify.com/v2/acts/accountable_eel~lever-job-delta-monitor/run-sync-get-dataset-items?token=<YOUR_TOKEN>, Body {"lever_company_id":["{{job posting}}"]}, mapping the row's job posting into the lever_company_id array.

MCP. In Claude, Cursor, or any MCP client with the Apify MCP server, ask for "Lever Job Delta Monitor | Apify" — the agent will find and run this actor.

Tips

• Filters do not read workplace_type. The keyword filter searches title, location and team only. To get remote roles, keep every row and filter the workplace_type column downstream.
• Group by team. If you send results to several people, split the dataset on team and each recipient only sees their own function.
• Give each audience its own deltaName. A newsletter and a coaching list should not share a baseline, or one list consumes the other's new roles.
• Check the slug in a browser first. Open jobs.lever.co/<slug>; if it loads, the actor will find it.
• The actor calls api.lever.co. That is Lever's main postings API. Companies hosted only on Lever's separate EU instance are not covered by this build, and we have not tested one.

vs. alternatives

	What it costs	What you get	Trade-off
This actor (lever-job-delta-monitor)	$0.001 per posting returned, $0.001 per run, nothing when no posting is new	New Lever postings with team, commitment and workplace type, tracked in your own store	Lever only, with no date-range or remote filter, and a first run that bills the full board unless seeded
Lever Jobs Lookup (lever-jobs-lookup)	Billed per company, listed at $0.004 per slug found	A full snapshot of each board with richer filters and an optional new-only mode	Better for a one-off pull or a short list. It is also cheaper than this actor whenever a board gains four or more postings between scheduled runs.
Lever's postings API	Free	The same JSON	You maintain the seen-list, the scheduling and the flattening yourself

The sibling's price is the one on its page at the time of writing; confirm it there.

FAQ

How do I find a company's Lever slug?

Look at the address of its careers page. If it is hosted on Lever, the slug follows jobs.lever.co/.

Does a quiet week cost anything?

Only the run start fee of $0.001. No new posting means no job charge.

What if a company reposts a role?

A repost normally carries its own id, so it comes back as a new row. Compare titles downstream if you want to drop reposts.

Can I see roles that were closed?

No. This monitor reports additions only.

Related actors

• Lever Jobs Lookup returns a full snapshot per company, for one-off research.
• ATS Jobs Unified Lookup identifies which applicant-tracking system a company runs.