This tutorial presents a design-automation-oriented introduction to machine learning methods for quantum circuit synthesis, mapping, and architecture search. Rather than treating quantum machine learning (QML) as a standalone application area, the tutorial frames QML, quantum reinforcement learning (QRL), and search-based optimization as algorithmic tools for quantum CAD under realistic hardware constraints, including native gate sets, limited qubit connectivity, circuit depth, noise, and resource overhead. Participants will learn how parameterized quantum circuits define structured design spaces, how reinforcement learning, evolutionary optimization, and differentiable architecture search can be used for design-space exploration, and how these methods interact with compilation, mapping, and benchmarking workflows. Through hands-on examples using open-source quantum simulators such as Qiskit and PennyLane in Google Colab, attendees will explore hardware-aware ansatz discovery and circuit optimization using practical design metrics such as fidelity, two-qubit gate count, depth, and robustness. The tutorial concludes with an outlook on scalable Quantum CAD, including emerging self-programming and automated quantum design frameworks.
