Filtering a US Job-Postings Dataset by Title, Location, Salary, and Skills

The core filters that define this dataset

Every record can be narrowed before download on job title, location, salary, required skills, seniority, contract type, and category, with employer available as a further cut for firmographic targeting.

These dimensions are not tags bolted on after the fact. The filter and segmentation builder is part of the delivered product, so the cut you define is the CSV you receive.

Because the filtering happens at the source, you skip the step of importing a whole-market file and writing your own logic to slice it down to the roles and places you care about.

How location is parsed into structured fields

Location is not left as a raw string. Each posting is parsed into separate city, state, zip, country, and region fields, so you can filter to a metro area or roll up to a wider region without cleaning free text yourself.

Structured location parsing is what makes local sourcing and regional roll-ups possible directly from the file, rather than after a separate import-and-clean step.

What the salary and compensation fields carry

Where a posting states pay, the record carries salary-range and employment-type fields, so compensation can be read alongside title and seniority rather than looked up in a second source.

Keeping pay in the same row as the role it belongs to is what lets the file double as a compensation reference for the roles it actually covers.

Why it ships as a dated CSV, not a live feed

This is a snapshot, not an API or streaming feed. Postings are filtered to the last 30 days and stamped with freshness timestamps, so every row reflects a recently active listing at the time of refresh.

Records arrive deduplicated across sources and documented by a data dictionary, so the file drops into a spreadsheet or database without integration work or a pipeline to maintain.

Role-family cuts on one shared schema

The whole-market cut is the US Live Job Postings CSV; from the same schema you can also take a single role family instead of the full market.

The engineering cut enriches each posting with AI-extracted required skills and technology tags, while the sales-manager cut applies a people-management filter and adds employer firmographics for territory targeting.

The registered-nurse cut classifies postings by healthcare employer and parses city, state, and zip for local sourcing; the financial-analyst cut adds firmographics, seniority, and O*NET occupation-family mapping; the construction-manager cut is geocoded by region.

When a team recruits across several of these at once, the five priority role-family files are available together on one shared schema and data dictionary.