Hacker News Search, Who-is-Hiring Jobs & Feeds

Search the entire Hacker News archive — stories and comments, back to 2007 — with no 1,000-result wall, turn the monthly "Ask HN: Who is hiring?" thread into structured job posts with contact e-mails, and pull the live front page. Built on Hacker News' official APIs: no key, no login, no proxies, no HTML scraping to break.

• 🔍 Unlimited-depth search — most HN scrapers stop at the API's 1,000-result cap; this one pages by time window instead (verified: 3,000 unique, correctly ordered rows in a single run)
• 💼 Who-is-hiring, parsed — the July 2026 thread returned 276 job posts: company 88.8%, e-mails on 26.8% of posts, URLs on 84.1%, remote flag on every row
• 💬 Comments are searchable too — where brand opinions actually live, not just titles
• ⚡ Fast & free to run — pure HTTP against official endpoints, no proxy or third-party cost

Quick start (API)

Every Hacker News mention of "supabase" in the last 90 days, stories and comments:

curl -X POST "https://api.apify.com/v2/acts/foxlabs~hackernews-intelligence/run-sync-get-dataset-items?token=YOUR_APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{ "mode": "search", "query": "supabase", "searchType": "both", "datePreset": "last_90_days", "maxResults": 500 }'

Prefer no code? Open the Input tab — the default is the latest Who-is-hiring thread — and click Start.

What you get

Search & feed rows:

Field	Type	Description
type	string	story, comment or job_post
title	string	Story title (comments carry their story's title)
url / hnUrl	string	The linked article + the Hacker News discussion page
author	string	HN username
points / numComments	number	Score and comment count (stories)
text	string	Story or comment body, HTML decoded to clean text
storyId / parentId	number	Thread position for comments
createdAt / fetchedAt	string	When it was posted / collected (ISO)

Who-is-hiring rows add:

Field	Type	Description
company	string	Employer, read from the thread's Company | Role | Location convention — null when a post doesn't follow it, never guessed
headline / roleLine	string	The post's first line, and the part after the company
remote	boolean	Whether the headline says remote
emails	array	Every e-mail address in the post (lowercased, de-duplicated)
urls	array	Application / company links in the post
text	string	The full post, clean text
threadId / threadTitle	number / string	Which monthly thread it came from

Sample output

Real rows from live runs (July 2026):

{
  "type": "job_post",
  "threadTitle": "Ask HN: Who is hiring? (July 2026)",
  "company": "Founding Technologist + Full Stack Developer",
  "headline": "Founding Technologist + Full Stack Developer | REMOTE | Part-or-full-time | $0 + equity",
  "roleLine": "REMOTE | Part-or-full-time | $0 + equity",
  "remote": true,
  "emails": ["hataginow@gmail.com"],
  "urls": [],
  "author": "baaaadegg",
  "hnUrl": "https://news.ycombinator.com/item?id=48919859",
  "createdAt": "2026-07-15T12:32:31Z"
}

{
  "type": "story",
  "id": 49063754,
  "title": "PGSimCity - How PostgreSQL Works",
  "url": "https://nikolays.github.io/PGSimCity/",
  "hnUrl": "https://news.ycombinator.com/item?id=49063754",
  "author": "jonbaer",
  "points": 920,
  "numComments": 91,
  "createdAt": "2026-07-27T00:19:04Z"
}

Download as JSON, CSV, Excel or HTML, or pull it straight from the API.

Modes & filters

• Mode — search (whole archive), who_is_hiring (monthly job thread), feed (live lists).
• Search — free-text query; search in stories, comments or both; minPoints / minComments to keep only stories that landed; author to follow one person; date presets from 24 hours to all time (plus custom).
• Sort — date (default) streams newest-first with unlimited depth; relevance uses the API's ranked order but is capped at 1,000 results by the API, and the run tells you so.
• Who is hiring — pick the thread type (who is hiring / who wants to be hired / freelancer) and any month (2026-07), or leave it empty for the latest.
• Feed — top / new / best / ask / show / jobs, optionally walking each comment tree (commentDepth, maxCommentsPerStory).
• Max results — hard cap; 0 really is unlimited.

Example inputs (copy & paste)

// 1) Brand monitoring — every mention, comments included
{ "mode": "search", "query": "clickhouse", "searchType": "both", "datePreset": "last_30_days" }

// 2) What actually landed — front-page-grade stories on a topic
{ "mode": "search", "query": "postgres", "minPoints": 100, "datePreset": "last_30_days" }

// 3) This month's job market, structured
{ "mode": "who_is_hiring", "threadType": "hiring", "maxResults": 0 }

// 4) Candidates instead of employers
{ "mode": "who_is_hiring", "threadType": "wants_to_be_hired" }

// 5) A specific month's hiring thread
{ "mode": "who_is_hiring", "month": "2026-04" }

// 6) Follow one writer
{ "mode": "search", "author": "patio11", "searchType": "both", "datePreset": "last_year" }

// 7) Today's front page with discussion
{ "mode": "feed", "feed": "top", "includeComments": true, "commentDepth": 2, "maxResults": 30 }

Use cases

• Developer-tool marketing. Track every mention of your product and your competitors in comments — the place where developers say what they really think — then schedule the run weekly and diff the results.
• Tech recruiting & job hunting. Turn the monthly Who-is-hiring thread into a spreadsheet: company, role line, remote flag, application e-mails and links. What used to be an hour of scrolling is one run.
• Launch & trend tracking. Watch Show HN and high-score stories in your category to see what is getting traction, with scores and comment counts attached.
• Research datasets. Pull a topic's full history — the archive goes back to 2007 and the sliding-window pagination really does walk all of it — for NLP, sentiment or trend analysis.
• Community & founder intelligence. Follow specific authors or threads, or sweep Ask HN for recurring pain points to mine for product ideas.

Performance & cost

Pure HTTP against Hacker News' official endpoints — no browser, no proxy, no third-party API costs. Measured on the platform (2026-07-30): a 3,000-row archive sweep took 73 seconds; the 276-post July hiring thread took 12 seconds; a filtered topic search returned in 4 seconds. Runs use a 512 MB container.

Integrations

JavaScript (apify-client):

import { ApifyClient } from 'apify-client';
const client = new ApifyClient({ token: 'YOUR_APIFY_TOKEN' });
const run = await client.actor('foxlabs/hackernews-intelligence').call({
  mode: 'who_is_hiring', threadType: 'hiring', maxResults: 0,
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();

Python (apify-client):

from apify_client import ApifyClient
client = ApifyClient("YOUR_APIFY_TOKEN")
run = client.actor("foxlabs/hackernews-intelligence").call(run_input={
    "mode": "search", "query": "rust", "searchType": "both", "datePreset": "last_30_days",
})
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item.get("title"), item.get("points"))

Also works with Make / n8n / Zapier, scheduled runs, webhooks, and the Apify MCP server for AI agents.

Data quality (measured 2026-07-30)

What	Result
Who-is-hiring parse (n=276, July 2026 thread)	company 88.8% · headline 100% · role line 89.1% · URLs 84.1% · e-mails 26.8% of posts · remote flag on every row
Verification against live HN pages	5 of 5 randomly picked job posts matched their HN page exactly (author + company)
Unlimited-depth proof	3,000 / 3,000 unique rows, all inside the requested window, strictly newest-first
Filter accuracy	minPoints=100 + 30-day window: 13/13 rows satisfied both

Nothing is fabricated. A field that a post doesn't contain comes back null or empty, never guessed.

Pricing

Pay per result — you're billed per row returned, and there's an Apify free tier to try it. The underlying APIs are free and public, so there are no proxy or third-party costs baked in.

FAQ

Where does the data come from? Two official Hacker News APIs: the Algolia HN Search API for the archive and the Firebase HN API for live feeds. Both are public and keyless.

How do you get past the 1,000-result limit? The search API caps classic pagination at 1,000 results. In date order this Actor doesn't use page numbers at all — it narrows a timestamp window on each round, so it can keep walking as deep as you ask. (Relevance sort still uses the ranked endpoint and its 1,000 cap; the run logs a warning when you pick it.)

Can I get the whole history of a keyword? Yes — set datePreset: "all_time" and maxResults: 0. HN's archive starts in 2007.

Why is company empty on some job posts? The hiring thread has a convention (Company | Role | Location) that most but not all posters follow — measured 88.8% on the July 2026 thread. When a post doesn't follow it we leave company null rather than guess; headline and the full text are always there.

Do I get e-mails for every job post? No — only where the poster included one, measured 26.8% of posts. The rest link to an application page (urls, 84.1%).

Is this legal? It reads public, official Hacker News APIs — no login, no bypassing anything. Contact details in hiring posts were published by the poster specifically to be contacted about that role; use them for that.

Can I search comments only? Yes, searchType: "comments" — useful for brand and competitor monitoring, since opinions live in the discussion rather than the title.

How fresh is it? Live. Both APIs reflect Hacker News in near real time; schedule the Actor to track a keyword or catch each new hiring thread.

Troubleshooting

• 0 results → widen the date window, drop minPoints, or simplify the query — the archive is searched exactly as you type it.
• Off-topic rows in search → the search API also matches a story's URL, so a keyword can hit a domain name. Add searchType: "stories" plus a minPoints filter to tighten it.
• "Capped at 1,000" warning → you're on relevance sort; switch Sort by to date for unlimited depth.
• No thread found in who-is-hiring mode → check the month format (YYYY-MM), or leave it empty to take the latest thread.

Notes, limits & legal

• Hacker News content belongs to its authors; this Actor redistributes what the official APIs publish. Not affiliated with Hacker News or Y Combinator.
• company parsing depends on a community convention (measured 88.8%) — the raw text is always included so you can re-parse it your own way.
• Relevance sort is limited to 1,000 results by the API; date sort is not.

Support

Questions, a field you'd like added, or a custom build? Open the Issues tab, or e-mail info@foxlabs.com.tr. We reply fast.

If this Actor saves you time, a ⭐ review really helps.

Changelog

0.2 — 2026-07-30

• Full-archive search (stories, comments or both) with points/comment/author filters and date presets back to 2007 — no 1,000-result wall in date order, thanks to sliding time-window pagination (verified: 3,000 unique ordered rows in one run).
• Who-is-hiring parser — the monthly thread (or wants to be hired / freelancer, any month) becomes structured job posts: company, headline, role line, remote flag, e-mails and URLs, plus full text. Measured on the July 2026 thread: 276 posts, company 88.8%, e-mails 26.8%, 5/5 verified against live HN pages.
• Feeds kept (top/new/best/ask/show/jobs, optional comment trees); HTML decoded to clean text; failures stay visible instead of ending as an empty "success".

0.1 — 2026-05-20

• Initial release: Firebase feeds with depth-limited comment-thread walking.

---

Part of the foXLabs data platform — official public-data company, contact, ownership, jobs, procurement & AI-search intelligence scrapers. Browse the full suite at data.foxlabs.com.tr.