---
license: cc-by-4.0
pretty_name: Zorky IT Job Market Aggregates
language:
- en
size_categories:
- 1K<n<10K
tags:
- jobs
- labor-market
- it-jobs
- tech-jobs
- salaries
- hiring
- technologies
- tabular
configs:
- config_name: roles
  data_files: roles.csv
  default: true
- config_name: countries
  data_files: countries.csv
- config_name: cities
  data_files: cities.csv
- config_name: roles_by_country
  data_files: roles_by_country.csv
- config_name: technologies
  data_files: technologies.csv
- config_name: technologies_by_role
  data_files: technologies_by_role.csv
- config_name: technologies_by_country
  data_files: technologies_by_country.csv
- config_name: salary_by_role
  data_files: salary_by_role.csv
- config_name: salary_by_technology
  data_files: salary_by_technology.csv
---

# Zorky IT Job Market Aggregates

Aggregated statistics on open IT vacancies indexed by [Zorky](https://zorky.tech) — demand by role,
technology, country and city, plus advertised salary ranges. Every number here is computed
from the public snapshot that powers [zorky.tech](https://zorky.tech); nothing in the dataset identifies
a single vacancy, employer or person.

**Data cut-off:** 2026-11-11T02:59:22Z · snapshot `zorky-snapshot-1-20261111T030955Z-39859a21`

## Please credit Zorky

If you use these numbers in an article, chart, report or model, please credit
**“Zorky (zorky.tech)”** and link to **https://zorky.tech** — ideally to the matching page from the
`zorky_url` column. That link is what keeps this dataset free.

```
Zorky (2026). Zorky IT Job Market Aggregates, snapshot 2026-11-11. https://zorky.tech
```

## At a glance

- **234,225** open vacancies in the public index (first seen 2026-04-26 – 2026-11-10).
- Largest roles: Fullstack (8.36%), Data (8.32%), DevOps (8.15%), Backend (7.93%), Architect (7.48%).
- Most mentioned technologies: Python (19.57%), AWS (15.11%), SQL (12.01%), Azure (11.59%), Java (11.04%).
- Largest countries: US (27.99%), PL (10.14%), GB (6.32%), IN (6.02%), DE (4.77%).
- Median advertised salary: **$8,436/month** across 67,211 vacancies with a salary.

## Files

| file | rows | contents |
|---|---:|---|
| `roles.csv` | 25 | Open vacancies by role (Zorky role classifier). |
| `countries.csv` | 108 | Open vacancies by country. |
| `cities.csv` | 2,326 | Open vacancies by city within country (vacancies without a recognisable city are left out). |
| `roles_by_country.csv` | 1,570 | Open vacancies by role within country. |
| `technologies.csv` | 53 | Open vacancies mentioning a technology (one vacancy can count for several). |
| `technologies_by_role.csv` | 1,126 | Technology x role. |
| `technologies_by_country.csv` | 2,840 | Technology x country. |
| `salary_by_role.csv` | 23 | Advertised monthly salary in USD by role, as published on zorky.tech/research/salaries. _(different universe — see Methodology)_ |
| `salary_by_technology.csv` | 51 | Advertised monthly salary in USD by technology, as on the technology pages of zorky.tech. _(vacancies of the public index with a normalised salary)_ |

`build_info.json` records the snapshot, thresholds, folding statistics and the exact builder version.

## Methodology

**Universe.** Open vacancies in Zorky's public job index at the cut-off — the same set that
[zorky.tech/jobs](https://zorky.tech/en/jobs) shows. A vacancy enters the index when it is open, passes
Zorky's quality gate (or was published by the employer), has a role assigned and either lists at
least two technologies or has a technical role. Freelance boards and hiring-signal channels are
excluded. Sources mix company career sites and applicant-tracking systems, job boards and
aggregators, and Telegram channels.

**Roles** come from Zorky's role classifier. `Other` and `IT` are catch-all buckets, not roles.

**Technologies.** 53 technologies that have their own page on
zorky.tech. A vacancy counts for a technology when the stack extracted from its text contains it;
one vacancy can count for several. Labels known to be unreliable are left out: `bun`, `creatio`, `defi`, `dynamics`, `go`, `jest`, `sui`, `vite`
(for example `go` also matches the English word).

**Geography.** Country is an ISO 3166-1 alpha-2 code as resolved by Zorky. City is normalised
(first segment of the location, common aliases such as Warszawa → Warsaw, “Remote” and bare country
names, regions and office labels dropped; so are employer names, job titles and placeholders that
sources put in the location field). 82,441 vacancies have no recognisable
city and are not in `cities.csv`.

**Salaries** are advertised salaries converted to USD per month; only vacancies with a known pay
period and 500–50,000 USD/month are used.
`salary_by_role.csv` is the published Zorky salary report: it covers **all open vacancies in the
Zorky database** with such a salary (67,211 vacancies), which is wider than the job index, and
keeps roles with at least 10 salaries and monotonic percentiles.
`salary_by_technology.csv` uses the vacancies of the job index; p25–p75 are shown from
10 salaries, p10 and p90 only from 25.

**Small groups.** Any group with fewer than **5** vacancies is folded into `(other)`.
If `(other)` itself would stay below 5, the smallest visible groups are folded into it as well,
so its value never reveals the size of one hidden group; a table that is below 5 as a whole is
not published. Two-way tables fold the outer group first, then the inner one.
With N = 5: 24 countries (41 vacancies),
6,197 city groups and 519 role × country cells are folded.

**Excluded on purpose:** vacancy identifiers, titles and texts; employer names; anything about
candidates; salary by country and by seniority — Zorky's own plausibility checks rejected both
(the country mix of salaried vacancies does not match the market, and the seniority field
measures how often it is filled rather than seniority).

## Limitations

- This is not a census of the labour market. Coverage depends on which sources Zorky reads, and
  the mix changes as sources are added. The largest countries in the index (US 27.99%, PL 10.14%, GB 6.32%, IN 6.02%) reflect
  that source mix as much as the market itself.
- This release has **no time series**. Zorky's weekly intake went from 24,940 vacancies (week of 2026-08-31) to 139,527 (week of 2026-11-02) as new sources were connected — a series of that kind measures our coverage, not the market. Comparable periodic releases of these tables are the intended way to show change over time.
- Role and technology labels are produced automatically and contain errors.
- Cities come from free-text locations: a state or region can appear where the source gives no city,
  and a vacancy listed in several places counts once, at the location given first.
- The same job cross-posted on several sites may be counted more than once.
- Only 11.3% of vacancies in the index state a usable salary; salaried vacancies skew towards
  markets where publishing pay is common or required.

## License

[Creative Commons Attribution 4.0 International](https://creativecommons.org/licenses/by/4.0/) (`CC-BY-4.0`).

Attribution required by the license: credit **“Zorky (zorky.tech)”** and link to **https://zorky.tech** — see “Please credit Zorky” above.

## Source and contact

Built from the public snapshot of [zorky.tech](https://zorky.tech) by `scripts/open_data/build_open_data.py`.
Questions and corrections: via [zorky.tech](https://zorky.tech).

## Кратко по-русски

Агрегаты по открытым IT-вакансиям из публичного снимка [zorky.tech](https://zorky.tech): спрос по ролям,
технологиям, странам и городам, вилки зарплат в USD/мес. Ни одной отдельной вакансии, компании или
человека; группы меньше 5 свёрнуты в `(other)`. Срез — 2026-11-11T02:59:22Z. При использовании —
ссылка на https://zorky.tech.
