Abstract: The proliferation of quantum computing system architectures and associated mechanisms for high-level programming has effectively introduced a stove-piped architecture that places a burden on programmers hoping to write quantum-classical applications that target a variety of quantum backends. Ultimately this is a compilation issue – how does one develop an efficient compilation approach that lowers any language representation to any quantum coprocessor architectural type? This problem arose in the classical computing world as well, and has been effectively solved through the introduction of a common intermediate representation (IR) (e.g. the LLVM) – an object model that language lowering methods can target with implementations provided for lowering the IR to available CPU instruction sets. Now researchers are seeking similar approaches for the introduction of a common quantum intermediate representation enabling a unified and cohesive software infrastructure for quantum-classical accelerated-node application development. We seek approaches that let quantum language compilers target a common IR layer that can be further lowered to any native backend architecture. The goal of this workshop is to provide a platform for discussing current state-of-the-art approaches to quantum intermediate representation development, expression, deployment, and utility. We will highlight recent approaches, future directions, and novel designs through a series of invited presentations on specific challenge topics from leaders in the field of quantum compilation and software. It is our hope that this workshop will spur collaboration and introduce the community to novel languages, compilers, and frameworks enabling a cohesive and unified software toolchain for quantum-classical computation.