zalizedata/us-work-visa-salary-dataset

US Work Visa & Salary Disclosures (H-1B, PERM, LCA)

14.5M salary disclosure records from official US Department of Labor and USCIS visa filings (H-1B LCAs, PERM) — employer, job title, wage and worksite, spanning FY2008 to the latest fiscal year.

Part of the DataForge Open Data program — full production packages, free for academic and personal use. Canonical dataset page: [https://data.zalize.com/datasets/us-work-visa-salary-dataset](https://data.zalize.com/datasets/us-work-visa-salary-dataset)

Formats & how to load

Native parquet (snappy) shards live under data/<table>/ and are rendered directly by the Dataset Viewer above. Original release zips (with CSV/JSONL copies, data dictionary, datasheet and QA report) are archived under archive/ and on <https://dl.zalize.com/open-data>.

# pandas
import pandas as pd
df = pd.read_parquet("hf://datasets/zalizedata/us-work-visa-salary-dataset/data/employer_profiles/employer_profiles-00000-of-00001.parquet")

# polars (lazy, all shards)
import polars as pl
lf = pl.scan_parquet("hf://datasets/zalizedata/us-work-visa-salary-dataset/data/employer_profiles/*.parquet")

# duckdb
import duckdb
rel = duckdb.sql("SELECT * FROM 'hf://datasets/zalizedata/us-work-visa-salary-dataset/data/employer_profiles/*.parquet' LIMIT 10")

Archived zips: visa_salaries-tier1-L-2026-08-02.zip, visa_salaries-tier1-M-2026-08-02.zip, visa_salaries-tier1-S-2026-08-02.zip

Packages in this repo

All files are also served from the machine-readable open-data index: <https://dl.zalize.com/open-data> (per-package URL: https://dl.zalize.com/open-data/<package_id>).

What you get

• —S — California snapshot: 1,814,772 disclosure records
• —M — Latest year, all programs: 915,005 records
• —L — Longitudinal full pack: 14,535,607 records (FY2008+)
• —Employer, job title, wage, worksite and case status fields
• —Data dictionary, datasheet and QA report included

Use cases

• —Compensation benchmarking
• —Talent-market research
• —Immigration analytics
• —Employer hiring intelligence

Source & methodology

US DOL OFLC disclosure files + USCIS H-1B data

Coverage, update cadence and the full field-level data dictionary are on the dataset page: https://data.zalize.com/datasets/us-work-visa-salary-dataset (DATA-DICTIONARY.md and DATASHEET.md are inside each package zip).

License

Dual license — CC BY-NC 4.0 (DataForge curated layer; academic/personal use, attribution + backlink required); commercial use requires a DataForge commercial license. Upstream data: public domain, U.S. federal government work (17 U.S.C. §105)

• —Academic / personal use: CC BY-NC 4.0 on the DataForge curated layer — attribution and a backlink to <https://data.zalize.com> are required.
• —Commercial use: requires a DataForge commercial license — contact us via the portal.
• —Upstream license terms continue to apply to the underlying data.

Citation

DataForge (data.zalize.com), built from official US DOL/USCIS disclosure files — https://data.zalize.com/datasets/us-work-visa-salary-dataset

DOI (Zenodo mirror): 10.5281/zenodo.21812582 · GitHub Release mirror: https://github.com/wookat/dataforge-pipelines/releases/tag/open-data-visa-salaries

Usage

from datasets import load_dataset

ds = load_dataset("zalizedata/us-work-visa-salary-dataset", "employer_profiles", split="train")
print(ds[0])

Available configs: employer_profiles, filings, uscis_h1b_employers, wage_benchmarks.

Related datasets

More DataForge open datasets in Jobs & Labor Market:

• —390,000+ Tech Job Postings with Parsed Salaries — Monthly Snapshot
• —tech-job-postings-salaries-sample

Full catalog (25 datasets): <https://data.zalize.com/open-data> · all HF repos: <https://huggingface.co/zalizedata>