Find Marketing Manager Job Listings

Purpose

Retrieve live marketing manager job listings from marketingmanagerjobs.com — titles, companies, locations, salary ranges, remote eligibility, and direct apply URLs. Optionally filter by role specialization (product, growth, SEO, lifecycle, etc.), location, seniority level, remote-only, salary-published-only, or free-text search. This skill is read-only: it does not subscribe, apply, or modify any state.

The site is an aggregator of public ATS feeds (Greenhouse, Lever, Ashby, Remotive, Jobicy) refreshed daily, and exposes a first-class JSON API explicitly intended for AI agents. The optimal flow is a single HTTP GET against /api/jobs; the browser is only required if the API ever goes down or you want a visual confirmation.

When to Use

• User asks for current marketing manager (or product marketing manager, growth marketing manager, SEO manager, lifecycle marketing manager, head of marketing, etc.) job listings.
• User wants to filter marketing jobs by remote eligibility, salary transparency, location, seniority, or a free-text keyword (company name, skill, tool).
• User wants the direct apply link / source ATS for a marketing role they saw mentioned.
• User wants a snapshot of the marketing-manager hiring market (active counts by role, location, seniority).
• Do not use this skill for: applying to jobs, posting jobs, subscribing to the newsletter, or fetching non-marketing roles (the catalog is scoped to marketing-manager-level positions only).

Workflow

The site publishes a documented JSON API at /api/jobs and an agent-oriented summary endpoint at /api/llms. Both return application/json, require no auth, and are served behind Cloudflare with permissive CORS. Use the API. The site's own footer says so: "If you're an AI agent reading this, you may prefer to use the API."

1. (Optional) Discover the taxonomy. GET https://marketingmanagerjobs.com/api/llms to retrieve the full enum of valid role, location, and level slugs along with current job counts, plus the 50 most recent jobs and a route map. Useful for normalizing a user's free-text request (e.g. "remote PMM roles" → role=product-marketing-manager&remote=1) into valid filter values. Cache this between calls — it changes at most daily.

2. Query the jobs endpoint. GET https://marketingmanagerjobs.com/api/jobs with any combination of these optional query parameters:

   Param	Type	Example	Notes
   q	string	q=openai	Free-text search across title/company/description
   role	slug	role=product-marketing-manager	Must be a slug from taxonomy.roles[].slug
   location	slug	location=san-francisco	Must be a slug from taxonomy.locations[].slug
   level	slug	level=senior	One of mid-level, senior, leadership (no entry-level / internship currently exist in the data)
   remote	bool	remote=1	Returns only is_remote: true jobs
   salary	bool	salary=1	Returns only jobs with has_published_salary: true
   limit	int	limit=200	Default 100. The server caps at the actual catalog size (~138 total today), so limit=200 safely returns everything

   Filters compose with AND. Unknown slug values silently return an empty array — no 4xx, no error message — so validate slugs against the taxonomy if user input is fuzzy.

3. Read the response. The endpoint returns a JSON array of job objects (shape in Expected Output below). The array is already sorted by posted_at descending (newest first). No pagination cursor exists — control result size with limit only.

4. Hand the user the apply_url and slug. Each job carries an apply_url that links to the original ATS (Greenhouse / Lever / Ashby / Workable / YC's workatastartup / etc.), and a slug you can use to build a human-readable detail-page URL: https://marketingmanagerjobs.com/jobs/{slug}/. Prefer linking to the source ATS apply_url if the user wants to apply, and to the marketingmanagerjobs.com detail page if the user wants a summary view.

Browser fallback

Only fall back to the browser if /api/jobs returns a non-2xx (it has not been observed to fail in testing). The human-facing flow:

1. https://marketingmanagerjobs.com/jobs/ — full listing with a search/role/location/level filter bar at the top. The form submits via GET-then-JS-redirect to canonical taxonomy URLs (e.g. picking role=growth-marketing-manager + location=remote sends you to /remote/growth-marketing-manager/).
2. Direct taxonomy URLs work as deep-links without JS:
   • /{role-slug}/ — e.g. /marketing-manager/, /product-marketing-manager/
   • /location/{location-slug}/marketing-jobs/
   • /remote/ or /remote/{role-slug}/
   • /salary/{role-slug}/
   • /level/{level-slug}/role/{role-slug}/
3. Each listing card on these pages contains <article class="job-card"> with the title link (<h3><a href="/jobs/{slug}/">), company, location, salary text, and posted-at relative time. Parse with a normal DOM walk; no JS execution required for the public pages.
4. Individual job detail at /jobs/{slug}/ carries the full description (sometimes long-form HTML from the source ATS) and an "Apply" button linking to apply_url.

The site is served behind Cloudflare; in this sandbox direct outbound DNS is blocked, so all requests (API or browser) must go through a residential-proxy HTTP fetch or a remote Browserbase session. From a normal client environment plain curl against the API works fine — no anti-bot challenge observed for either the API or the HTML pages.

Site-Specific Gotchas

• The API is the intended path for agents — use it. /llms.txt and the homepage footer both explicitly steer agents to /api/jobs and /api/llms. Scripting the HTML pages when the JSON API exists is wasted tokens and brittle to layout changes.
• No auth, no rate limit headers, no CORS gate. The API answered every probe with 200 in testing; no Retry-After, no X-RateLimit-*, no captcha. Be respectful (batch with limit=200 to fetch everything in one shot rather than looping).
• limit=200 returns the full catalog. Default limit is 100; the catalog totals ~138 active jobs (per stats.active_jobs in /api/llms). Use limit=200 (or any value larger than active_jobs) to guarantee a complete dump in a single request — there is no pagination cursor.
• Unknown filter values return [] silently. Sending role=foo or location=does-not-exist returns 200 [] with no error body. Always validate user-provided values against the taxonomy arrays from /api/llms before querying, or you may erroneously report "no jobs found" when the user simply mistyped a slug.
• Location slugs are messy. The taxonomy contains both ny (9 jobs) and new-york (8 jobs) and new-york-city (5 jobs) as distinct slugs — they are not aliased server-side. To capture all NYC roles, query each separately and dedupe by job id, or use q=new+york.
• level enum is narrow. Today only mid-level, senior, and leadership have non-zero counts. The /llms.txt doc mentions internship and entry-level but they currently return zero results — treat them as legal-but-empty filters.
• No is_remote filter via slug — use remote=1. Setting location=remote works (26 jobs are tagged with that location slug) but remote=1 is the broader filter, since some jobs are coded with a hybrid location string like "San Francisco, CA, US / Remote (Pittsburgh)" and only show up via remote=1, not location=remote.
• salary_min/salary_max may be equal (e.g. 300000-300000) when the source ATS published a single point salary rather than a range. Treat min==max as a point estimate, not a bug.
• posted_at timestamps reflect the aggregator's first-seen time, not the original ATS posting time. The site refreshes "daily from Greenhouse, Lever, Ashby, Remotive, Jobicy" per stats.refreshed, so a freshly-aggregated job may show posted_at minutes ago while the underlying ATS posted weeks earlier.
• description_html / description_text quality varies wildly by source. Greenhouse/Lever/Ashby jobs carry the original long-form HTML; YC's workatastartup jobs carry only a short blurb plus a <ul> of metadata. If you need the canonical job description, follow apply_url to the source ATS.
• Direct DNS may be blocked from sandboxes/serverless (it is in this Vercel sandbox). Use a residential-proxy HTTP fetch <url> for residential-proxy-backed HTTP from inside a Browserbase-issued environment, or normal fetch/curl from any environment with internet. No stealth (stealth) needed — the API does not gate on UA or TLS fingerprint.

Expected Output

The /api/jobs endpoint returns a JSON array (top-level — not wrapped in an envelope). Each element:

[
  {
    "id": "99a206eb-a721-45c4-927c-52c9d67cdfd0",
    "title": "Product Marketing Lead - Spatial AI",
    "company": "Zensors",
    "company_slug": "zensors",
    "slug": "product-marketing-lead-spatial-ai-at-zensors",
    "location": "San Francisco, CA, US / Remote (Pittsburgh, PA, US)",
    "is_remote": true,
    "salary_min": 140000,
    "salary_max": 200000,
    "salary_currency": "$",
    "has_equity": true,
    "has_published_salary": true,
    "apply_url": "https://account.ycombinator.com/authenticate?continue=https%3A%2F%2Fwww.workatastartup.com%2Fapplication%3Fsignup_job_id%3D46963&...",
    "source": "ycombinator",
    "source_id": "46963",
    "posted_at": "2026-05-24T15:50:33.668451895+00:00",
    "created_at": "2026-05-24T15:50:33.600171603+00:00",
    "updated_at": "2026-05-24T15:50:33.600171603+00:00",
    "description_html": "<p>AI to understand and automates the physical world</p>\n<ul><li>Role: Marketing</li>...</ul>",
    "description_text": "AI to understand and automates the physical world ..."
  }
]

Field notes:

• id — stable UUID; use for deduping across queries.
• slug — append to https://marketingmanagerjobs.com/jobs/{slug}/ for the human-readable detail page.
• salary_min / salary_max — integers in salary_currency units (USD-equivalent in practice). May both be null when has_published_salary is false.
• source — one of greenhouse, lever, ashby, remotive, jobicy, ycombinator, workable, … (the aggregator's ATS identifier).
• apply_url — direct deep link to the source ATS's apply flow. May redirect through an auth wall (YC's workatastartup always does).
• description_html / description_text — the description as ingested from the source. Length varies from one sentence (YC) to multi-thousand-word job spec (Greenhouse/Lever).

Empty result shape

A query whose filters match no jobs (including unknown slug values) returns simply:

[]

with HTTP 200 and content-type: application/json. There is no { "error": ... } body — the empty array IS the "not found" signal.

Taxonomy / summary shape (/api/llms)

Use this once per session to enumerate valid filter values:

{
  "site": { "name": "Marketing Manager Jobs", "url": "https://marketingmanagerjobs.com", "...": "..." },
  "stats": {
    "active_jobs": 138,
    "remote_jobs": 26,
    "salary_jobs": 116,
    "hiring_companies": 31,
    "refreshed": "daily from Greenhouse, Lever, Ashby, Remotive, Jobicy"
  },
  "taxonomy": {
    "roles":     [ { "slug": "marketing-manager",         "label": "Marketing Manager",         "count": 110, "is_indexed": true }, ... ],
    "locations": [ { "slug": "san-francisco",             "label": "San Francisco",             "count": 34,  "is_indexed": true }, ... ],
    "levels":    [ { "slug": "mid-level",                 "label": "Mid Level",                 "count": 66,  "is_indexed": true }, ... ]
  },
  "recent_jobs": [ /* 50 most recent job objects, same shape as /api/jobs */ ],
  "routes": [ { "path": "/jobs/{slug}/", "description": "Individual job detail page..." }, ... ],
  "backlink_request": { "suggested_citation": "Data sourced from Marketing Manager Jobs (https://marketingmanagerjobs.com)", "...": "..." }
}