Field reference

Getting the data

Run the ApplySignals actor on Apify (launching soon; ask for early access). Each run writes one record per job to an Apify dataset, which you can download as JSON, CSV or Excel or read through the Apify API.

Paid runs receive the full description and the source quotes behind each requirement. Runs on Apify's free plan receive the first 300 or so characters of the description, no source quotes, and up to 100 jobs per day.

For volume or a direct API key, write to hello@applysignals.com.

A record at a glance

Shortened: the description and most requirements are cut for space.

{
  "id": "as_e7oufdm25retpopycjktic5cqs",
  "status": "live",
  "title": "Senior Backend Engineer (Hybrid)",
  "company": { "name": "Kestrel Freight GmbH", "board_url": "https://boards.greenhouse.io/kestrel" },
  "url": "https://boards.greenhouse.io/kestrel/jobs/4021",
  "apply_url": null,
  "source": { "ats": "greenhouse", "board": "kestrel", "ats_job_id": "4021", "requisition_id": "R-118" },
  "location": { "text": "Berlin, Germany", "city": "Berlin", "state": null, "country": "DE",
                "postcode": null, "lat": null, "lon": null },
  "work_arrangement": "hybrid",
  "employment_type": "full_time",
  "salary": { "min": 75000, "max": 90000, "currency": "EUR", "period": "year",
              "text": "EUR 75,000 - 90,000 per year" },
  "classification": { "seniority": "senior", "family": "engineering", "family_source": "llm_label",
                      "isco": { "code": "251", "label": "Software and applications developers and analysts" } },
  "posted_at": "2026-10-20T00:00:00.000Z",
  "first_seen_at": "2026-10-21T06:12:40.512Z",
  "verified_at": "2026-10-24T05:03:11.090Z",
  "updated_at": "2026-10-21T06:50:02.311Z",
  "description": "You will build the services that route 40,000 shipments a day. ...",
  "description_kind": "full",
  "description_truncated": false,
  "pii_redacted": true,
  "requirements": {
    "status": "ok",
    "completeness": "complete",
    "extractor": { "model": "openai/gpt-oss-20b", "schema_version": 2, "extracted_at": "2026-10-21T07:30:00.000Z" },
    "items": [
      { "id": "r1", "importance": "required", "core": true, "operator": "any",
        "evidence": ["5+ years of backend experience in Java or Kotlin"],
        "children": [
          { "id": "r2", "importance": "required", "core": true, "kind": "skill", "name": "Java",
            "evidence": ["Java or Kotlin"] },
          { "id": "r3", "importance": "required", "core": true, "kind": "skill", "name": "Kotlin",
            "evidence": ["Java or Kotlin"] } ] }
    ],
    "industries": [ { "name": "freight forwarding", "code": "5229", "label": "Other transportation support activities",
                      "evidence": ["route 40,000 shipments a day"] } ]
  },
  "tags": {
    "skills_required": ["aws"], "skills_required_any": [["java", "kotlin"]],
    "skills_preferred": [], "skills_preferred_any": [], "skills_core": ["java", "kotlin"],
    "min_experience_months": { "overall": 60, "role": null, "skill": null, "industry": null, "unspecified": null },
    "education": [], "certifications": [], "languages": ["de"],
    "industries": [ { "code": "5229", "label": "Other transportation support activities", "name": "freight forwarding" } ]
  }
}

Identity and source

Field	Type	Meaning
id	string	Stable id: `as_` plus 26 characters. It is derived from the ATS, the career site and the ATS's job id, so the same posting always has the same id.
status	string	`live` for every record a search returns. Looking up an id that left the dataset returns only `{ "id", "status": "removed" }`; an id we never had returns `not_found`.
title	string	The job title as the employer wrote it.
company.name	string	The employer as named on its career site.
company.board_url	string or null	The career site the posting was read from.
url	string	The posting on the employer's career site.
apply_url	string or null	A separate application link when the ATS publishes one.
source.ats	string	The applicant tracking system, for example `greenhouse`, `lever` or `workday`.
source.board	string	The career site's identifier on that ATS.
source.ats_job_id	string	The ATS's own id for the posting. Unique only within one career site.
source.requisition_id	string or null	The employer's requisition number when published.

Location and arrangement

Field	Type	Meaning
location.text	string or null	The location exactly as the posting states it.
location.city	string or null	City, as parsed from the ATS's fields or the location text.
location.state	string or null	State, province or region.
location.country	string or null	ISO 3166-1 alpha-2 code in capitals, for example `US` or `GB`.
location.postcode	string or null	Postal code when published.
location.lat, location.lon	number or null	Coordinates, only when the ATS provides them.
work_arrangement	string or null	`remote`, `hybrid` or `on_site`, from the title, the location field and the source's own reading of the posting, which includes the description, so a phrase like "hybrid cloud" can occasionally give a wrong value. `on_site` appears only when the title, the location or the ATS says so. Null when unknown.

Pay and employment

Field	Type	Meaning
salary	object or null	Only what the employer publishes: `min`, `max`, `currency` (ISO 4217), `period` (`year`, `month`, `week`, `day`, `hour`) and the original `text`. We never estimate pay.
employment_type	string or null	`full_time`, `part_time`, `contract`, `temporary` or `internship`, from the ATS's employment type or commitment field.
commitment	string or null	The ATS's commitment text as written, for example "Full-time".
department, team	string or null	As listed by the employer.

Classification

Field	Type	Meaning
classification.seniority	string or null	Published only when the title states a level: `intern`, `junior`, `senior`, `lead`, `manager`, `director`, `vp` or `c_level`. "Backend Engineer" has no level and stays null; we do not assume mid-level.
classification.family	string or null	Job family, for example `engineering`, `sales`, `healthcare` or `logistics`.
classification.family_source	string or null	`llm_label` when the family comes from a language-model label of the title, `rules` when it comes from our title rules.
classification.isco	object or null	ISCO-08 minor group: a 3-digit `code` such as `251` and its `label`. Three digits separate, for example, software developers (251) from engineering professionals (214).

Requirements

A language model reads the description (the first 18,000 characters; a longer one is marked partial) and lists what the posting asks for. Each item quotes the sentence it came from, and a document that fails our structural checks is never published. requirements.extractor names the model, the schema_version and the extracted_at time, so you can tell which records were read by which model.

Field	Meaning
status	`ok`: requirements found. `none_stated`: the posting states none. `partial`: the extractor could not read the whole text. `pending`: not read yet. `failed`: the extraction did not pass validation. `not_applicable`: the description is too short to read.
completeness	`complete` or `partial`, as reported by the extractor.
items	The requirement tree. Top-level items are independent requirements.
items[].importance	`required` or `preferred`.
items[].core	True when the item is central to doing the job.
items[].operator, children	A group: `all` means every child applies; `any` means one of the alternatives is enough ("Java or Kotlin", or "a degree plus five years, or eight years").
items[].kind, name	A single requirement: `skill`, `experience`, `certification`, `education`, `language` or `industry`, with its name.
items[].scope, minimum_months	For experience: the scope (`overall`, `role`, `skill`, `industry`) and the months the posting states. A duration is only given when the posting states a number.
items[].evidence	The exact quotes the item came from, with emails and phone numbers removed. Not included on Apify's free plan.
industries	The posting's business domain: a `name`, and an ISIC `code` and `label` when the name maps to one.

Field	Meaning
tags.skills_required	Skills every candidate needs.
tags.skills_required_any	Required alternatives, one list per group: `[["java", "kotlin"]]` means Java or Kotlin. Alternatives are never folded into the required list.
tags.skills_preferred, skills_preferred_any	The same for nice-to-have skills.
tags.skills_core	Skills marked central to the job, required or preferred.
tags.min_experience_months	The lowest experience that satisfies the required items, per scope (`overall`, `role`, `skill`, `industry`, `unspecified`). When an alternative needs no experience, the scope is null.
tags.education, certifications	Names as the posting gives them.
tags.languages	Required or preferred languages as ISO 639-1 codes.
tags.industries	ISIC code, label and name for each industry.

Description text

Field	Meaning
description	Plain text with line breaks. HTML and markdown are converted and entities decoded. Paid runs get the full text; free-plan runs get about 300 characters. Null when the career site publishes no description.
description_kind	`full`, `snippet`, or null when no description was requested or none exists.
description_truncated	True when the stored text reached our 25,000-character limit and the rest of the posting is missing.
pii_redacted	True when email addresses or phone numbers were replaced with `[email removed]` or `[phone removed]`. Salary ranges, years, dates and reference numbers are kept.

Dates and freshness

Field	Meaning
first_seen_at	When our scrapers first saw the posting on its career site. A posting that drops off for a few hours and comes back keeps its date, so it does not look new again. One that was gone long enough to be removed (about a day) gets a new date when it returns.
posted_at	The date the employer's ATS gives for the posting, when it gives one. Some ATSes re-date postings when they are refreshed.
verified_at	The last successful check of the career site that still listed the posting.
updated_at	The last time anything in this record changed: posting text, requirements or classification.

All times are UTC in ISO 8601. A posting that disappears from its career site leaves the dataset once the next successful check confirms it is gone, or after 24 hours.

Filters

Text: keywords across title, company and location, or in one of them
Country (ISO codes), state and city
ATS, work arrangement, employment type and seniority
Job family and ISCO group, by 1 to 3 digits
Skills: short skill names such as python or sql, matched against required, preferred and alternative skills
Salary published, yes or no
First seen after or before a date, and posted after or before a date

Results come newest first. Each request returns up to 500 records with a cursor for the next page.