Can I train machine learning models on job posting data?

Yes, TheirStack sells job, company and technology datasets in bulk for training and evaluating models. Job descriptions come with structured fields such as title, seniority, location, salary and technologies, which make useful labels.

In which formats are the TheirStack datasets delivered?

TheirStack delivers its datasets as Parquet or CSV files through S3, with history and daily or hourly updates. You request temporary credentials from the API and download the files with your usual tools.

How much history do the job datasets include?

The TheirStack job datasets include postings back to 2021, including jobs that have since closed. That history lets you train and test models on how roles and skills change over time.

What does a job posting dataset contain?

A TheirStack job posting dataset contains each job's title, description, company, location, seniority, salary when stated, the technologies it mentions and the date it was posted. The company and technology datasets add each employer's firmographics and tech stack.

Power model training with job and company dataNew

Train, fine-tune or evaluate your models on years of structured job ads and company records, delivered as files.

Who it's for: An AI lab, model developer or data-science team that needs large volumes of structured B2B text.

Try it free See integrations

ByteDance posted a Machine Learning Engineer role · 2h ago
Alibaba opened AI research roles in Singapore · 6h ago
Spotify is hiring for its personalization models · 1d ago

How model training with job data works

TheirStack delivers job, company and technology datasets with postings back to 2021 as files, so you train, fine-tune and evaluate your models on structured job ads.

Download

Download the history as files

TheirStack delivers its job, company and technology datasets as Parquet or CSV files through S3, with postings back to 2021.

Datasets
Parquet or CSV through S3
Postings back to 2021

Training rows

Company · files in S3Posted

ByteDanceMachine Learning Engineer, senior, PyTorch2024

AlibabaAI Research Scientist, staff, JAX2023

SpotifyMachine Learning Engineer, mid level, TensorFlow2022

Company · files in S3Posted

ByteDanceMachine Learning Engineer, senior, PyTorch2024

AlibabaAI Research Scientist, staff, JAX2023

SpotifyMachine Learning Engineer, mid level, TensorFlow2022

Label

Use the structured fields as labels

TheirStack gives each job description its title, seniority, location, salary when stated and the technologies it mentions, ready to use as labels.

Job data fields
Seniority and salary per job
Technologies found in the text

Labelled rows

CompanySeniority

ByteDanceMachine Learning Engineer, PyTorchSenior

AlibabaAI Research Scientist, JAXStaff

SpotifyMachine Learning Engineer, TensorFlowMid level

Update

Add new postings when you retrain

TheirStack adds new files every day or hour, including the jobs that have since closed, so each training run starts from current data.

New files every day or hour
Closed jobs kept in the history
Company and technology datasets too

Files since your last run

CompanyAdded

ByteDanceMachine Learning Engineer, new postingToday

AlibabaResearch Scientist, closedToday

SpotifyData Scientist, new postingYesterday

Learn how

Docs
Datasets
Access complete jobs, closed jobs, technographics, and company datasets — delivered as Parquet or CSV, with historical coverage and daily or hourly updates via S3.

Power model training with job and company data: integrations with AI assistants, automations and CRMs

TheirStack works with the AI assistants, automation tools and CRMs you already use: ask Claude or ChatGPT through the MCP server, send new matches to Clay, n8n, Make or Zapier by webhook, and create leads in Salesforce or HubSpot.

AI assistants
Ask Claude, ChatGPT, Cursor or any MCP client to search jobs, companies and tech stacks through the TheirStack MCP server.
MCP server
https://api.theirstack.com/mcp
Read the MCP docs
Automations
TheirStack sends every new matching company by webhook as soon as it finds it, to your automation platform or through it into your CRM and sales tools.
Automation platforms
CRMs and sales tools
Run this use case from your own code
```
curl -X POST https://api.theirstack.com/v1/jobs/search \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "job_country_code_or": ["US"],
    "posted_at_max_age_days": 730
  }'
```
Read the API docs Read the webhooks docs