Synthetic dataset of data science job postings modelled on HiringCafe

Simulated dataset generated for Practice 1 of the course M2.851 Tipología y ciclo de vida de los datos(Master's in Data Science, Universitat Oberta de Catalunya), November 2026. The real dataset was scraped from https://hiringcafe.com/ and cannot be redistributed under the site’sTerms and Conditions. This synthetic version keeps the exact structure of the real one (same files,columns, types and serialisation) but contains no real content: identifiers, URLs, company names,taglines, descriptions and locations are generated with the Faker library. Categorical and numericcolumns (seniority, workplace type, tools, degree requirements, salary, company sector and size…) aresampled row-wise from the real dataset so as to have realistic distributions. Files job_postings_dataset.csv: 1,000 rows, 38 columns, one row per job posting. Key: job_id. hiring_orgs_dataset.csv: 300 rows, 19 columns, one row per organisation. Key: org_id,referenced in the job_postings_dataset.csv as a foreign key. Lists are stored inside a cell separated by |. The composite fields salary and locations areStored as the text representation of a Python dict. Empty cells indicate the attribute wasn't available. License CC BY-SA 4.0. Authors: Santiago Tomás Nadal, Miguel Sierra Arroyo.