Weekly data note. 890 job postings published on Bumeran (bumeran.com.ar), 2 to 9 November 2026, collected with an open extractor and published as open data. Numbers, not opinions.
The week in numbers
- 890 postings collected, one row per listing, de-duplicated by URL.
- Work mode: 2 % fully remote, 15 % hybrid, 83 % on site.
- Salary transparency: Salaries are almost never disclosed on Bumeran: fewer than 20 postings this week carried a figure, so no salary statistics are reported.
Where the jobs are
| City | Postings |
|---|---|
| Capital Federal | 340 |
| Buenos Aires | 29 |
| Vicente López | 27 |
| Pilar | 22 |
| San Martin | 21 |
| Rosario | 20 |
| Córdoba | 20 |
| Avellaneda | 14 |
| La Plata | 12 |
| Martínez | 12 |
Contract types
| Contract | Postings |
|---|---|
| full time | 838 |
| part time | 45 |
| not stated | 4 |
| internship | 3 |
Who is hiring the most
| Employer | Postings |
|---|---|
| ADECCO RECURSOS HUMANOS ARGENTINA SA | 33 |
| Grupo Gestión | 24 |
| Randstad Argentina | 21 |
| Umano HR | 18 |
| CONA CONSULTORES EN RRHH | 15 |
| RH Talentum | 13 |
| BAYTON | 12 |
| MJ Consultores | 10 |
| Bumeran Selecta | 8 |
| ACCIONA IT | 8 |
Employers are companies and public bodies as named on the listing; confidential listings are excluded.
How to read this
- These are advertised postings, not hires. A posting can be renewed or duplicated by an agency; we de-duplicate by URL only.
- The sample is what Bumeran published over seven days and what the public pages expose; it is not a random sample of the whole Argentina labour market.
- Labels (sector, contract, work mode) are the site's own, normalised to the schema.org JobPosting vocabulary.
Get the data yourself
The full week is published as open data (ODC-By) on Hugging Face and refreshed every Monday. The collection runs on a public Apify Actor, Bumeran & LatAm Jobs Scraper: any keyword, city or search URL, JSON or CSV out, first results in under a minute. AI agents can query the same sources through the Job Boards MCP Server.
Generated by an AI agent from the dataset described above, with no human editing.