Research and learning

Causal inference, decisions and experimentation

This family of work connects predictions to decisions. It uses causal and experimental reasoning to show what a model, feature or policy actually changes, so AI companies can act on evidence rather than on correlation.

AI products now shape what users see, what they buy and how systems behave, so teams need to know what their changes cause. Predictive lift alone does not establish treatment effects. These hires are tricky because the titles blur. Good candidates usually come from statistics, econometrics, experimentation or decision science, but many have only run simple tests. Look for people who state their assumptions, explain where an estimate could fail, and tie the analysis to a real choice.

Hiring now 4

Roles in this family

4 roles

Where the people come from

Statistics, econometrics, experimentation or decision science

Not to be confused with. Predictive lift alone does not establish treatment effects.

Core skills

  • Probability and statistics
  • Experimental design
  • Causal inference
  • Decision theory and game theory

Where it is used

  • Cross-sector AI platforms
  • Banking and payments
  • Marketing advertising and sales

More in Research and learning

← All new job families

Start a search

Hiring in causal inference, decisions and experimentation?

Tell us the work you need done. We find the people already doing it, whatever their job title says.