Hacker News "Who Is Hiring" Jobs Dataset

What’s in the file

20 columns over 93,339 rows. Distinct counts and null rates are measured from the file itself, not estimated.

Things worth noticing

• TypeScript went from 0.3% of ads in 2012–14 to 21.8% in 2024–26, and React from 0.9% to 23.3%. Over the same span Ruby fell from 17.1% to 3.8% and Rails from 15.6% to 4.1% — the clearest before-and-after in the file, and a fair summary of what happened to the web stack.
• Python is the one constant. It appears in 22.9% of ads in 2012–14 and 23.4% in 2024–26, holding its share almost exactly while everything around it turned over. Rust and LLM tooling go the other way, from literally zero mentions to 7.4% and 11.0%.
• Remote peaked and then gave ground. 11.9% of ads mentioned remote in 2011 and 29.3% in 2019; that hit 77.5% in 2021 and has fallen every year since, to 51.8% in 2026 — still four times the pre-pandemic norm, and well off the peak.
• Hiring volume tracks the funding cycle with no lag: 10,278 ads in 2021, then 4,171 in 2023 — a 59% drop in two years. Europe held up better than the US through it, rising from 11.5% of located ads in 2011 to 19.2% in 2026 while the US share fell from 63.6% to 50.1%.

Put an agent on it

Anything you were about to do with this file by hand is a job that runs every week forever. Each of these is a working prompt — open it and it is already loaded.

On the first working day of each month, read the newest "Ask HN: Who is hiring?" thread and compare it against the trailing twelve months in this dataset.

Report:
- Total ads this month against the 12-month average, as a count and a percentage
- The five technologies that gained the most share, and the five that lost the most, each with both percentages
- Remote share this month against the same month last year
- Any role family whose share moved more than 3 percentage points
- The split between United States and Europe, and whether it moved

Write it as six sentences with the numbers inline. Lead with the single biggest move. If the month is unremarkable, say that in one line rather than manufacturing a narrative.

Track this list of companies across the monthly Hacker News hiring threads: [list them here].

Each month, for any that posted, report:
- Which roles they advertised, and how that differs from their previous posts
- Any technology that appears in their ad for the first time, or has disappeared from it
- Any change in location or remote policy
- How often they have posted over the last two years, and whether the rate is rising or falling

Flag explicitly when a company that posted consistently stops, and when one starts advertising a role family it never has before — a first ML or infrastructure hire says more about direction than a press release does.

I will name a technology, for example "Rust" or "Kubernetes".

From this dataset, build its adoption curve: share of all ads mentioning it, by month, from its first appearance to now. Then:
- Identify the month it crossed 1% and the month it crossed 5%, if it has
- Compare its first 36 months against the first 36 months of two or three comparable technologies, aligned to each one's own start rather than to calendar time
- Split the current share by region and by role family
- Say whether the trailing 12 months are rising, flat or falling, with the numbers

Finish with three sentences on whether the curve looks like early adoption, a plateau, or a decline — and note where the Hacker News skew towards startups is likely to overstate the case.

Load it yourself

The file is a plain parquet. Nothing here is specific to us.

-- No download step: DuckDB reads the parquet over HTTP range requests.
INSTALL httpfs;
LOAD httpfs;

CREATE VIEW hacker_news_who_is_hiring AS
SELECT * FROM read_parquet('https://cotera.co/datasets/hacker-news-who-is-hiring.parquet');

SELECT * FROM hacker_news_who_is_hiring ORDER BY post_id LIMIT 20;

More datasets