Kanvaz AI Labs
All open roles

Core R&D / Physical AI

Research Engineer: Causal Discovery & Sim-to-Real

About the role

Kanvaz is bridging the gap between theoretical causal inference and real-world physical systems. We are looking for a Research Engineer to lead the design and implementation of active causal discovery algorithms on real hardware. You will work at the intersection of structural causal models (SCMs) and physical experimentation, moving beyond synthetic data to validate interventions on actual physical testbeds.

Target background: MSc/PhD in Statistics, Mathematics, or Computer Science.

What you'll do

  • Design and execute experiments to validate causal discovery algorithms under real-world physical constraints.
  • Develop and maintain physical testbeds (akin to the Causal Chambers initiative) to stress-test SCM learning.
  • Formulate greedy learning strategies for Structural Causal Models (SCMs) that are robust to hardware noise and latency.
  • Collaborate with the simulation team to close the "sim-to-real" gap, ensuring algorithms trained in simulation perform reliably on physical hardware.

What you bring

  • Strong background in Causal Inference and Structural Causal Models (SCMs).
  • Experience with active learning and intervention design in physical or simulated environments.
  • Proficiency in Python and statistical computing; experience with hardware integration (sensors/actuators) is a plus.
  • You are a problem-solver who thrives on the friction between theoretical algorithms and messy physical reality.

Apply