New Jobs

Time window for job listings. Smaller windows return fresher jobs in smaller batches; larger windows are best for backfills.

• 1h - firehose of just-discovered jobs (best for hourly polling).
• 24h - most common choice for daily syncs.
• 7d - weekly batch jobs.
• 6m - full backfill (very large; description search (description, description_advanced) is allowed on 6m but can be slow and may time out - simplify the query, lower limit, or use a shorter time window if so).

Number of results to return. Default: 100, min: 1, max: 1000.

Use with offset to paginate, e.g. limit=100&offset=0, then limit=100&offset=100, then limit=100&offset=200. For very large feeds prefer cursor over offset.

Number of results to skip for pagination. e.g. limit=100&offset=200 returns the third page (results 201–300). Use cursor instead for very large feeds - offset becomes inefficient at deep pages.

Cursor-based pagination. Pass the last job's id to get the next page. Note: When cursor is set, results are ordered by id ascending - this overrides the default ordering (date_posted desc, or date_modified desc on /modified-ats). Cursor and offset pagination therefore traverse the feed in different orders.

Interaction with offset: cursor overrides offset. If both are passed, offset is ignored. Pick one pagination strategy and stick with it.

Format the job description should be returned in. Choose text for plain text or html for raw HTML. Omit the parameter entirely to exclude the description field from the response - descriptions are not included by default because they significantly increase payload size.

Caution: html returns the description as it was scraped from the source - sanitize before rendering on a web page.

Off by default to keep responses small. Set to true to include basic LinkedIn-derived organization fields inline with each job: org_linkedin_name, org_linkedin_website, org_linkedin_industry, org_linkedin_headcount, org_linkedin_size, org_linkedin_followers, org_linkedin_headquarters, org_linkedin_locations, org_linkedin_type, org_linkedin_founded_date, org_linkedin_slogan, org_linkedin_description, org_linkedin_specialties, org_linkedin_recruitment_agency_derived, org_crunchbase_categories, org_crunchbase_total_investment, org_logo_permalink. org_linkedin_slug is always returned (regardless of this flag) so jobs can be joined to /organizations-advanced.

You don't need this flag if you're already calling /organizations-advanced - every field above is included there (alongside additional advanced fields).

Full-text search on job title (weighted, natural language). Searches like a Google query:

• software engineer - jobs with both words in the title.
• "software engineer" - exact phrase.
• software OR engineer - either word.
• -engineer - exclude jobs with engineer in the title.

For Boolean operators, parentheses, or prefix matching, use title_advanced instead.

Full-text search on job description (weighted, natural language). Same syntax as title. e.g. python OR javascript OR js returns descriptions mentioning any of those keywords. Heavy queries (long phrases, many ORs) on long time windows can time out when using time_frame=6m - keep it tight.

Full-text search on location fields (weighted, natural language). Search using full names, abbreviations are not supported.

• Use United States, not US.
• Use United Kingdom, not UK.
• US states need their full name and country: New York, United States, California, United States.
• UK cities need the constituent country: Birmingham, England, United Kingdom, Glasgow, Scotland, United Kingdom.

Multi-location with OR: "United States" OR "United Kingdom" or "San Francisco" OR "New York". For Boolean operators (parentheses, AND, NOT), use location_advanced.

See Nuances of Location Search for more.

Filter by organization name(s). Exact, case-sensitive match - nvidia will not match NVIDIA. For fuzzy or Boolean matching, use organization_advanced. Caveat: organization names containing parentheses are not supported here - use organization_slug (preferred when available) or organization_advanced instead.

Fetch the full list of organizations we currently track via /organizations.

Input a comma-separated list.

Exclude organization name(s). Same matching rules as organization (exact, case-sensitive). For fuzzy/Boolean exclusion, use organization_advanced with !.

Input a comma-separated list.

Boolean full-text search on title. Operators: & (AND), | (OR), ! (NOT), <-> (followed-by), parentheses for grouping, :* for prefix matching.

Multi-word phrases must be single-quoted ('machine learning') or chained with <-> (machine <-> learning). If title is also passed, title_advanced takes precedence and title is ignored.

Examples:

• (AI | 'machine learning' | robotics) & !marketing - AI/ML/robotics roles, no marketing titles.
• senior & (python | rust) & !staff - senior IC roles in either stack, excluding staff-engineer.
• project <-> manag:* - matches project manager, project management, project managing, ...

See Advanced Searching Guide for more.

Boolean full-text search on description. Same operator syntax as title_advanced. If description is also passed, description_advanced takes precedence and description is ignored. Description searches are expensive when using time_frame=6m, if a query times out, simplify it, lower limit, or shrink the time window.

Boolean full-text search on location. Same operator syntax as title_advanced. If location is also passed, location_advanced takes precedence and location is ignored. The same naming rules as location apply (full names, no abbreviations).

Examples:

• 'United States' & !California - US minus California.
• Germany & !(Berlin | Munich) - Germany excluding two cities.
• 'New York' | 'San Francisco' | Boston - multi-city.

Boolean full-text search on organization name. Same operator syntax as title_advanced. If organization is also passed, organization_advanced takes precedence and organization is ignored.

Examples:

• University & !Harvard - every "University" employer except Harvard.
• (Microsoft | Google | Apple) - match any of the three.
• Open <-> AI:* - matches OpenAI, Open AI, Open-AI, etc.

Filter jobs posted on or after this date (ISO 8601). e.g. 2026-04-20T00:00:00. Posted-date timestamps are UTC, with a 1-2 hour ingestion delay between the source publishing the job and it appearing in this API.

See Time Fields for the difference between date_posted, date_created, and date_modified.

Filter jobs posted before this date (ISO 8601).

Filter jobs indexed on or after this date (ISO 8601). date_created reflects when our system first indexed the job.

Filter jobs indexed before this date (ISO 8601).

Filter by employer domain(s). The domain_derived field is mapped to jobs using an LLM agent with 99% accuracy. Please report any mistakes.

This parameter is especially useful to capture employers that have multiple ATS across regions. For example, nvidia.com,microsoft.com,google.com. Each tracked organization's domain is exposed via /organizations.

Input a comma-separated list.

Exclude employer domain(s). Same rules as domain. e.g. nvidia.com,microsoft.com.

Input a comma-separated list.

Filter by source ATS/job board(s). Currently supported values: adp, applicantpro, ashby, bamboohr, breezy, careerplug, comeet, csod, dayforce, dover, eightfold, firststage, freshteam, gem, gohire, greenhouse, hibob, hirebridge, hirehive, hireology, hiringthing, icims, isolved, jazzhr, jobvite, join.com, kula, lever.co, manatal, oraclecloud, pageup, paradox, paycom, paycor, paylocity, personio, phenompeople, pinpoint, polymer, recruitee, recooty, rippling, rival, smartrecruiters, successfactors, taleo, teamtailor, trakstar, trinet, ultipro, werecruit, workable, workday, zoho. New ATS platforms are added regularly, recheck this list periodically.

Input a comma-separated list.

Exclude source ATS/job board(s). See source for the full list of supported values.

Input a comma-separated list.

Filter by LinkedIn industry. Use the exact LinkedIn industry name (case-sensitive). Common values: Software Development, Information Technology and Services, Financial Services, Hospital & Health Care, Staffing and Recruiting, Marketing and Advertising. The full taxonomy of ~400 industries lives at LinkedIn Industries.

For names containing & (e.g. "Hospital & Health Care"), .

Input a comma-separated list.

Exclude LinkedIn industry/industries. Same matching rules as organization_industry (exact, case-sensitive). See LinkedIn Industries for the full taxonomy.

Input a comma-separated list.

Filter by LinkedIn organization slug(s), the company-specific portion of a LinkedIn URL.

For https://www.linkedin.com/company/walmart/, the slug is walmart. For https://www.linkedin.com/company/tesla-motors/, it's tesla-motors.

Exact match only. Generally more reliable than organization (no spelling/casing variation) and the only way to filter on names containing parentheses.

Fetch the full list of organizations (with their slugs) via /organizations.

Input a comma-separated list.

Exclude LinkedIn organization slug(s). Same rules as organization_slug (exact match). Fetch slugs via /organizations.

Input a comma-separated list.

Filter companies with at least this many LinkedIn employees (inclusive lower bound). Source: LinkedIn company profile (org_linkedin_headcount).

Combine with organization_headcount_lt for a range, e.g. organization_headcount_gte=50&organization_headcount_lt=500 for 50–499-employee companies.

Filter companies with fewer than this many LinkedIn employees (exclusive upper bound). Source: LinkedIn company profile (org_linkedin_headcount).

Combine with organization_headcount_gte for a range.

Filter by AI-derived experience level (years of experience required, inferred from the job description and other elements from the job page).

• 0-2
• 2-5
• 5-10
• 10+

Pass multiple to broaden the match (e.g. 0-2,2-5). Input a comma-separated list.

Filter by AI-derived work arrangement.

• On-site - no remote at all.
• Hybrid - split between office and remote.
• Remote OK - fully remote, but an office is available.
• Remote Solely - fully remote, no office.

Pass multiple to broaden, e.g. Hybrid,Remote OK,Remote Solely for any-remote-flexibility roles. Input a comma-separated list.

Filter by AI-derived employment type(s) (overlap match, a job tagged with multiple types matches if any are in your list)

• FULL_TIME, PART_TIME
• CONTRACTOR, TEMPORARY, PER_DIEM
• INTERN, VOLUNTEER, OTHER

Input a comma-separated list.

Filter by AI-derived education requirements. The LLM is instructured to flag education that's required, preferred, or desired (overlap match, jobs are tagged with one or more requirements).

• high school
• associate degree
• bachelor degree
• professional certificate
• postgraduate degree

Note: Overlap match means bachelor degree matches jobs that explicitly require a bachelor's, it doesn't broaden to also match jobs that only require high school.

Input a comma-separated list.

Filter by AI taxonomy/taxonomies. Similar to ai_taxonomies_a, but only matches the first (most relevant) taxonomy assigned to a job. Use when you want a tight category match, e.g. jobs that are primarily about Technology, not just adjacent to it.

Input a comma-separated list.

Filter by AI taxonomy/taxonomies (overlap match, jobs are tagged with one or more, ordered by relevance). The first taxonomy in the array is the primary taxonomy.

Matches a job if any of its assigned taxonomies are in your list (broad). For tighter matches use ai_taxonomies_a_primary (matches only the primary). To exclude taxonomies use exclude_ai_taxonomies_a.

For names containing & (e.g. "Management & Leadership"), please double-quote.

Input a comma-separated list.

Exclude AI taxonomy/taxonomies - excludes a job if any of its assigned taxonomies are in this list. See ai_taxonomies_a for the value list and overlap-match semantics.

Input a comma-separated list.

Filter listings by AI-derived visa-sponsorship availability. Omit to include both.

• only - return only listings that offer visa sponsorship.
• exclude - exclude listings that offer visa sponsorship (return only those that don't).

Note: Detection is based on language in the job description, not all companies are very clear about their requirements in the job description.

Set to true to only return jobs that have salary information (raw or AI-enriched). Omit (or pass false) to include both.

Filter listings by recruitment-agency status of the organization. Omit to include both.

• only - return only listings whose organization is a recruitment/staffing agency.
• exclude - exclude listings from recruitment/staffing agencies.

The agency flag itself is exposed in the response as org_linkedin_recruitment_agency_derived.

Filter by one or more internal Fantastic.jobs IDs. Matches the id field in each response object, the same identifier surfaced by /expired-ats and /modified-ats for expiration and change tracking. e.g. id=8123456,8123457 returns those two specific jobs (useful for refreshing rows you already store).

Input a comma-separated list.

The Authorization header is used to authenticate with the API using your API key. Value is of the format Bearer YOUR_KEY_HERE.

Time window for job listings. Smaller windows return fresher jobs in smaller batches; larger windows are best for backfills.

• 1h - firehose of just-discovered jobs (best for hourly polling).
• 24h - most common choice for daily syncs.
• 7d - weekly batch jobs.
• 6m - full backfill (large; description search (description, description_advanced) is not available on 6m for /active-jb).

Number of results to return. Default: 100, min: 1, max: 1000.

Use with offset to paginate, e.g. limit=100&offset=0, then limit=100&offset=100, then limit=100&offset=200. For very large feeds prefer cursor over offset.

Number of results to skip for pagination. e.g. limit=100&offset=200 returns the third page (results 201–300). Use cursor instead for very large feeds - offset becomes inefficient at deep pages.

Cursor-based pagination. Pass the last job's id to get the next page. Note: When cursor is set, results are ordered by id ascending - this overrides the default ordering (date_posted desc, or date_modified desc on /modified-ats). Cursor and offset pagination therefore traverse the feed in different orders.

Interaction with offset: cursor overrides offset. If both are passed, offset is ignored. Pick one pagination strategy and stick with it.

Filter listings by recruitment-agency status of the organization. Omit to include both.

• only - return only listings whose organization is a recruitment/staffing agency.
• exclude - exclude listings from recruitment/staffing agencies.

The agency flag itself is exposed in the response as org_linkedin_recruitment_agency_derived.

Heads-up: AI-enriched fields (every ai_* response field) are not populated for agency listings on this endpoint, see the endpoint description.

Format the job description should be returned in. Choose text for plain text or html for raw HTML. Omit the parameter entirely to exclude the description field from the response - descriptions are not included by default because they significantly increase payload size.

Caution: html returns the description as it was scraped from the source - sanitize before rendering on a web page.

Set to true to exclude recruiter fields (recruiter_name, recruiter_title, recruiter_url) from the response. Default: included.

Full-text search on job title (weighted, natural language). Searches like a Google query:

• software engineer - jobs with both words in the title.
• "software engineer" - exact phrase.
• software OR engineer - either word.
• -engineer - exclude jobs with engineer in the title.

For Boolean operators, parentheses, or prefix matching, use title_advanced instead.

Full-text search on job description (weighted, natural language). Same syntax as title. e.g. python OR javascript OR js returns descriptions mentioning any of those keywords. Not available for time_frame=6m. Heavy queries (long phrases, many ORs) on long time windows can time out - keep it tight.

Full-text search on location fields (weighted, natural language). Search using full names, abbreviations are not supported.

• Use United States, not US.
• Use United Kingdom, not UK.
• US states need their full name and country: New York, United States, California, United States.
• UK cities need the constituent country: Birmingham, England, United Kingdom, Glasgow, Scotland, United Kingdom.

Multi-location with OR: "United States" OR "United Kingdom" or "San Francisco" OR "New York". For Boolean operators (parentheses, AND, NOT), use location_advanced.

See Nuances of Location Search for more.

Filter by organization name(s). Exact, case-sensitive match - nvidia will not match NVIDIA. For fuzzy or Boolean matching, use organization_advanced. Caveat: organization names containing parentheses are not supported here - use organization_slug (preferred when available) or organization_advanced instead.

Fetch the full list of organizations we currently track via /organizations.

Input a comma-separated list.

Exclude organization name(s). Same matching rules as organization (exact, case-sensitive). For fuzzy/Boolean exclusion, use organization_advanced with !.

Input a comma-separated list.

Boolean full-text search on title. Operators: & (AND), | (OR), ! (NOT), <-> (followed-by), parentheses for grouping, :* for prefix matching.

Multi-word phrases must be single-quoted ('machine learning') or chained with <-> (machine <-> learning). If title is also passed, title_advanced takes precedence and title is ignored.

Examples:

• (AI | 'machine learning' | robotics) & !marketing - AI/ML/robotics roles, no marketing titles.
• senior & (python | rust) & !staff - senior IC roles in either stack, excluding staff-engineer.
• project <-> manag:* - matches project manager, project management, project managing, ...

See Advanced Searching Guide for more.

Boolean full-text search on description. Same operator syntax as title_advanced. If description is also passed, description_advanced takes precedence and description is ignored. Not available for time_frame=6m. Description searches are expensive, if a query times out, simplify it, lower limit, or shrink the time window.

Boolean full-text search on location. Same operator syntax as title_advanced. If location is also passed, location_advanced takes precedence and location is ignored. The same naming rules as location apply (full names, no abbreviations).

Examples:

• 'United States' & !California - US minus California.
• Germany & !(Berlin | Munich) - Germany excluding two cities.
• 'New York' | 'San Francisco' | Boston - multi-city.

Boolean full-text search on organization name. Same operator syntax as title_advanced. If organization is also passed, organization_advanced takes precedence and organization is ignored.

Examples:

• University & !Harvard - every "University" employer except Harvard.
• (Microsoft | Google | Apple) - match any of the three.
• Open <-> AI:* - matches OpenAI, Open AI, Open-AI, etc.

Filter jobs posted on or after this date (ISO 8601). e.g. 2026-04-20T00:00:00. Posted-date timestamps are UTC, with a 1-2 hour ingestion delay between the source publishing the job and it appearing in this API.

See Time Fields for the difference between date_posted, date_created, and date_modified.

Filter jobs posted before this date (ISO 8601).

Filter jobs indexed on or after this date (ISO 8601). date_created reflects when our system first indexed the job.

Filter jobs indexed before this date (ISO 8601).

Filter by source job board(s). Input a comma-separated list.

Note on expiry tracking: only linkedin listings are re-checked for expiration and surfaced via /expired-jb. wellfound and ycombinator listings stay in the active feed until naturally aged out, so plan your freshness logic accordingly if you ingest them.

Exclude source job board(s). Input a comma-separated list.

Filter by LinkedIn-posted seniority level. English values: Internship, Entry level, Associate, Mid-Senior level, Director, Executive, Not Applicable. Case-sensitive.

Heads-up: many employers default to Not Applicable. If you filter to a specific level, you'll miss those, consider whether that's acceptable. Non-English postings may use localized seniority strings; if you target multiple markets, include both the English and local-language values.

Alternative: for a normalized, market-independent signal, use ai_experience_level instead (note: AI fields exclude staffing-agency listings, see the endpoint description).

Input a comma-separated list.

Filter by LinkedIn industry. Use the exact LinkedIn industry name (case-sensitive). Common values: Software Development, Information Technology and Services, Financial Services, Hospital & Health Care, Staffing and Recruiting, Marketing and Advertising. The full taxonomy of ~400 industries lives at LinkedIn Industries.

For names containing & (e.g. "Hospital & Health Care"), please double-quote.

Input a comma-separated list.

Exclude LinkedIn industry/industries. Same matching rules as organization_industry (exact, case-sensitive). See LinkedIn Industries for the full taxonomy.

Input a comma-separated list.

Filter by LinkedIn organization slug(s), the company-specific portion of a LinkedIn URL.

For https://www.linkedin.com/company/walmart/, the slug is walmart. For https://www.linkedin.com/company/tesla-motors/, it's tesla-motors.

Exact match only. Generally more reliable than organization (no spelling/casing variation) and the only way to filter on names containing parentheses.

Fetch the full list of organizations (with their slugs) via /organizations.

Input a comma-separated list.

Exclude LinkedIn organization slug(s). Same rules as organization_slug (exact match). Fetch slugs via /organizations.

Input a comma-separated list.

Filter companies with at least this many LinkedIn employees (inclusive lower bound). Source: LinkedIn company profile (org_linkedin_headcount).

Combine with organization_headcount_lt for a range, e.g. organization_headcount_gte=50&organization_headcount_lt=500 for 50–499-employee companies.

Filter companies with fewer than this many LinkedIn employees (exclusive upper bound). Source: LinkedIn company profile (org_linkedin_headcount).

Combine with organization_headcount_gte for a range.

Filter by raw employment type(s) (overlap match, a job tagged with multiple types matches if any are in your list). Values are country-specific and vary by job board.

Alternative: for normalized, market-independent values use ai_employment_type (note: AI fields exclude staffing-agency listings, see the endpoint description).

Input a comma-separated list.

Filter by AI-derived employment type(s) (overlap match, a job tagged with multiple types matches if any are in your list).

• FULL_TIME, PART_TIME
• CONTRACTOR, TEMPORARY, PER_DIEM
• INTERN, VOLUNTEER, OTHER

For raw, country-specific values use employment_type. Like every ai_* filter on /active-jb, this excludes staffing-agency listings, see the endpoint description.

Input a comma-separated list.

Filter by AI-derived experience level (years of experience required, inferred from the job description and other elements from the job page).

• 0-2
• 2-5
• 5-10
• 10+

Pass multiple to broaden the match (e.g. 0-2,2-5). Input a comma-separated list.

Filter by AI-derived work arrangement.

• On-site - no remote at all.
• Hybrid - split between office and remote.
• Remote OK - fully remote, but an office is available.
• Remote Solely - fully remote, no office.

Pass multiple to broaden, e.g. Hybrid,Remote OK,Remote Solely for any-remote-flexibility roles. Input a comma-separated list.

Filter by AI-derived education requirements. The LLM is instructured to flag education that's required, preferred, or desired (overlap match, jobs are tagged with one or more requirements).

• high school
• associate degree
• bachelor degree
• professional certificate
• postgraduate degree

Note: Overlap match means bachelor degree matches jobs that explicitly require a bachelor's, it doesn't broaden to also match jobs that only require high school.

Input a comma-separated list.

Filter by AI taxonomy/taxonomies. Similar to ai_taxonomies_a, but only matches the first (most relevant) taxonomy assigned to a job. Use when you want a tight category match, e.g. jobs that are primarily about Technology, not just adjacent to it.

Input a comma-separated list.

Filter by AI taxonomy/taxonomies (overlap match, jobs are tagged with one or more, ordered by relevance). The first taxonomy in the array is the primary taxonomy.

Matches a job if any of its assigned taxonomies are in your list (broad). For tighter matches use ai_taxonomies_a_primary (matches only the primary). To exclude taxonomies use exclude_ai_taxonomies_a.

For names containing & (e.g. "Management & Leadership"), please double-quote.

Input a comma-separated list.

Exclude AI taxonomy/taxonomies - excludes a job if any of its assigned taxonomies are in this list. See ai_taxonomies_a for the value list and overlap-match semantics.

Input a comma-separated list.

Filter listings by AI-derived visa-sponsorship availability. Omit to include both.

• only - return only listings that offer visa sponsorship.
• exclude - exclude listings that offer visa sponsorship (return only those that don't).

Note: Detection is based on language in the job description, not all companies are very clear about their requirements in the job description.

Set to true to only return jobs that have salary information (raw or AI-enriched). Omit (or pass false) to include both.

Filter listings by direct-apply availability. "Direct apply" means the candidate can apply within the source job board itself (e.g. LinkedIn's Easy Apply / one-click in-platform application) rather than being redirected to an external site. Omit to include both.

• only - return only listings that support direct apply on the source job board.
• exclude - exclude direct-apply listings (return only those that redirect to an external apply flow).

Direct-apply listings have almost no overlap with the ATS feed (/active-ats), making them complementary rather than duplicates.

Use this only if you also consume /active-ats - when true it removes the majority of LinkedIn jobs we've matched against an existing ATS posting. Omit (or pass false) to include both.

How matching works. For each LinkedIn job we run two checks against the ATS dataset:

1. Job title + organization name match.
2. Job title + LinkedIn company-profile match.

If either check hits, the job is flagged ats_duplicate=true. If neither hits, it's ats_duplicate=false. Some jobs are not checked - agency listings (org_linkedin_recruitment_agency_derived=true) and direct-apply listings (direct_apply=true) - and are flagged ats_duplicate=null (these are also kept when this filter is true, since they're unlikely to be ATS duplicates by construction).

We aim for high precision over recall, so a small number of duplicates can still slip through. For full deduplication, add a fuzzy-match layer client-side.

Filter by one or more LinkedIn job posting IDs (comma-separated when passed as a raw URL). This is LinkedIn's public job ID - the digits at the end of a linkedin.com/jobs/view/{linkedin_id} URL - surfaced as linkedin_id in each response object. Distinct from id, which is the internal Fantastic.jobs ID.

Filter by one or more internal Fantastic.jobs IDs. Matches the id field in each response object, the same identifier surfaced by /expired-jb for expiration tracking. e.g. id=8123456,8123457 returns those two specific jobs (useful for refreshing rows you already store). Distinct from linkedin_id, which is LinkedIn's public posting ID.

Input a comma-separated list.

The Authorization header is used to authenticate with the API using your API key. Value is of the format Bearer YOUR_KEY_HERE.