About the role
Explore learned compilation policies that choose how to transform a quantum circuit for different hardware targets. Our goal is to move beyond a single fixed sequence of compiler passes and investigate adaptive decisions informed by circuit structure, hardware topology and performance data.
This role connects machine learning research with compiler engineering, careful evaluation and reproducible experiments.
In this role, you will
- Prototype models and policies for compiler pass selection, routing or hardware-aware optimization.
- Build training and evaluation datasets from circuit and compilation trajectories.
- Compare learned approaches with established compilation baselines, including their compute costs.
- Design experiments that measure generalization across circuits, backends and changing hardware characteristics.
What we are looking for
- Strong Python experience and knowledge of modern machine learning workflows.
- Experience with reinforcement learning, graph learning, sequence modeling or combinatorial optimization.
- Ability to design rigorous evaluations and communicate experimental results.
Also valuable
- Familiarity with quantum compilation, compiler intermediate representations or hardware connectivity graphs.