Kanvaz AI Labs
All open roles

Foundational Research

Machine Learning Scientist: Invariance & Distribution Shift

About the role

Kanvaz operates in dynamic, non-stationary environments. We are looking for a Scientist to tackle the challenge of distribution shifts and invariance. Your work will focus on ensuring that our structural causal models remain robust when the underlying physical environment changes, using techniques like instrumental variable estimation and invariant risk minimization.

Target background: MSc/PhD in Machine Learning.

What you'll do

  • Research and implement methods for Invariant Causal Prediction to ensure model reliability under distribution shift.
  • Develop algorithms for Instrumental Variable (IV) estimation to handle confounding in observational data.
  • Work on representation learning techniques that isolate causal mechanisms from spurious correlations in physical sensor data.
  • Publish findings and contribute to the open-source community surrounding causal machine learning.

What you bring

  • Expertise in Causality, Invariance, and Distribution Shift.
  • Familiarity with Instrumental Variables and Structural Equation Models.
  • A research mindset with a focus on applied machine learning.
  • Ability to bridge the gap between theoretical ML (e.g., invariance work) and practical implementation.

Apply