Hacker News Who Is Hiring Monitor & Scraper

Watch the official Hacker News monthly threads (Ask HN: Who is hiring?, Ask HN: Who wants to be hired? and Ask HN: Freelancer? Seeking freelancer?) and get only the new posts matching your keywords since your last run.

The "Who is hiring?" thread gets hundreds of high-quality job posts spread across the whole month. Everyone reads it on day 1, then stops checking — but a third of the posts land later, when almost nobody is looking. Watching the tail is the edge, and nobody re-reads a 600-comment thread every two days. This monitor does.

Who uses this HN hiring monitor

• Job seekers and freelancers: schedule it every 1-2 days with your stack keywords and answer late posts while the thread is quiet again.
• Recruiters and founders: switch threadType to wants-to-be-hired and get the candidates who published their availability this month.
• Agencies: threadType: freelancer turns the monthly gig thread into a lead feed.
• Market watchers: keyword on a technology and count how often it shows up month over month.

How this Hacker News Who is hiring scraper works

1. Finds the latest official monthly thread of the type you choose (posted by the whoishiring account — imitator threads are excluded).
2. Reads every top-level post and keeps the ones matching your keywords (e.g. remote, react, europe, video).
3. Remembers every post it has already seen (per thread type + keywords, in a persistent state store) and outputs only the NEW ones on every later run. When a new month's thread starts, it rolls over automatically.

Input example

{
  "threadType": "hiring",
  "keywords": ["remote", "typescript"]
}

The defaults work as-is. Pipe the dataset to Slack or email through an Apify integration — each item's firstLine is a ready-made notification — and keep one monitor per keyword set, since each set gets its own memory through its monitor key.

Output example

{
  "postId": "48748100",
  "postUrl": "https://news.ycombinator.com/item?id=48748100",
  "firstLine": "Paris, France | VideoLab | Senior Video Editor (freelance welcome) | Remote EU",
  "text": "Paris, France | VideoLab | Senior Video Editor (freelance welcome) | Remote EU\nWe need a creative editor for AI-generated brand films. Contact: jobs@…",
  "postedAt": "2026-07-01T16:00:00.000Z",
  "threadTitle": "Ask HN: Who is hiring? (July 2026)",
  "matchedKeywords": ["remote"],
  "isNewSinceLastRun": true
}

A SUMMARY record in the run's key-value store gives the counts: posts in thread, matching, new, total known.

Data returned for every HN post

Field	What it holds
postId, postUrl	HN item id and direct link to the comment
firstLine	The Company | Role | Location header line, ready to use as a notification
text	Full post text, including whatever contact details the poster published
postedAt	When the post was added to the thread
threadTitle	Which monthly thread it came from
matchedKeywords	Which of your keywords hit
isNewSinceLastRun	Whether this run is the first to see the post

The three thread types

• hiring — companies posting jobs. For job seekers and freelancers hunting warm openings.
• wants-to-be-hired — candidates posting availability. For recruiters and founders sourcing talent.
• freelancer — the monthly gig-matching thread, both "seeking work" and "seeking freelancer" posts. For agencies and independents.

FAQ

What happens at month rollover? The monitor always targets the latest official thread; new month = new posts = they appear as new. Memory of past posts is kept 60 days, so nothing is double-reported.

First run floods me. That's the baseline (flagged isBaseline: true); disable includeBaselineInOutput to start silent.

How do I reset the memory? Set a new monitorId.

Fair use & data

• Reads only Hacker News' public API (Algolia). No login, no scraping.
• Poster usernames are deliberately dropped — the post text already contains whatever contact info the company chose to publish.
• Replies under posts are not collected; only the top-level job/availability posts.