Tech Stack Hiring Signals Monitor: Greenhouse, Lever & Ashby

Find out what technologies a company is hiring for right now, and how that differs from what it was hiring for before. Point the Actor at a list of companies and it reads their public job boards, classifies every open posting by technology stack, and tells you which stacks are emerging, rising or fading: on the very first run. Schedule it and it also reports every posting opened or closed and every stack-count change since the previous run.

No login, no cookies, no LinkedIn. It uses the free public job-board APIs that companies on Greenhouse, Lever and Ashby expose on purpose.

Who this is for

• SaaS and dev-tool sales teams: a company that starts hiring React engineers after years of Angular postings, or posts three "LLM engineer" roles in a month, is a buying signal. Get it before the competition.
• Recruiters and staffing agencies: see which stacks are heating up at target accounts before the reqs hit LinkedIn.
• Competitive and market intelligence: track a portfolio of competitors' technology bets over time.
• Investors and analysts: hiring shifts are a leading indicator of product direction.

What you get

One row per company:

Field	Meaning
alerts	Actionable findings above your threshold: emerging (stack appears in recent postings, none in older ones), rising (recent share ≥ 2× older share), fading (present in older postings, absent from recent), and after the first run new-since-last-run / gone-since-last-run. Each has a level (high/medium) and a plain-English detail.
counts	Open postings total, how many are recent vs older, how many are undated.
stacks	Every detected technology with total / recent / older counts and how many times it appears in a job title (a much stronger signal than a mention in the body).
topRecentStacks, titleStacks	Quick-glance summaries.
openedSincePrevious, closedSincePrevious, stackDeltasSincePrevious	Changes since the last run of this Actor for the same company.
boardUrl, ats	Where the data came from.

Turn on Include one row per posting to also get every open posting with its detected stacks, age, department, location and URL.

The key-value store holds a SUMMARY record with the run totals and every alert in one place, handy for a Slack or email integration.

How the analysis works

1. Resolve the board. A bare slug (stripe) is probed on Greenhouse, then Lever, then Ashby. A board URL (boards.greenhouse.io/<slug>, jobs.lever.co/<slug>, jobs.ashbyhq.com/<slug>) pins the ATS. Companies on other systems (Workday, iCIMS, SuccessFactors, LinkedIn-only) are reported as not-found and are not charged.
2. Classify. Every posting's title and description are matched against a built-in dictionary of ~55 technologies across frontend, languages, mobile, cloud, infrastructure, data, AI/ML, platforms and APIs, with the obvious traps handled ("Go to market" is not Go, "Reactive" is not React, "rust belt" is not Rust, Java is not JavaScript). Add your own keywords with Extra stack keywords.
3. Split by age. Every ATS publishes a posting date, so postings younger than the Recent window (default 90 days) form the "recent" set and older open postings form the baseline. That is what lets the Actor detect a shift on the first run, without waiting to accumulate history.
4. Diff. With Compare with previous run on (default), a compact snapshot per company is kept in a named key-value store, and each run reports what changed.

Input

{
  "companies": ["stripe", "https://jobs.lever.co/palantir", "https://jobs.ashbyhq.com/notion"],
  "recentDays": 90,
  "alertThreshold": 3,
  "extraStacks": ["Hasura", "Temporal"],
  "includePostingRows": false,
  "compareWithPreviousRun": true
}

An alert needs both alertThreshold postings and at least minSharePercent (default 3%) of the postings in its window, so a 600-posting board doesn't fire on 3 stray mentions while a 20-posting startup still does. Lower alertThreshold to 1 or 2 for tiny companies, or set minSharePercent to 0 to use the absolute threshold alone.

Pricing

Pay per event: a small charge per company analyzed plus a per-run start fee. Companies that cannot be found, or whose board has no public postings, are reported for free. Set a max charge on the run and the Actor analyzes as many companies as that allows and tells you which were skipped.

Limitations

• Coverage is limited to companies whose careers site runs on Greenhouse, Lever or Ashby. That is a large share of tech companies and startups, but not the Fortune 500 on Workday.
• The classifier counts mentions, not sentiment. A React posting that says "migrating away from Angular" still counts as one Angular mention. Title mentions (titleStacks) are the cleaner signal.
• Slugs are usually the company name but not always; if a company comes back not-found, open its careers page, find the board URL, and paste that instead.
• Only listed, public postings are visible. Confidential or internal reqs are not.

Related

Built by Anywhere Web Repair. See also our Website SEO Audit, EDGAR Filings Monitor and FDA Recalls & Enforcement Monitor.