High Assurance Software for Financial Regulation and Business Platforms

The financial technology sector is undergoing a transformation in moving to open-source and collaborative approaches as it works to address increasing compliance and assurance needs in its software stacks. Programming languages and validation technologies are a foundational part of this change. Based on this viewpoint, a consortium of leaders from Morgan Stanley and Goldman Sachs, researchers at Microsoft Research, and University College London, with support from the Fintech Open Source Foundation (FINOS) engaged to build an open programming stack to address these challenges.

The resulting stack, Morphir, centers around a converged core intermediate representation (IR), MorphirIR, that is a suitable target for existing languages in use in major investment banks and that is amenable to analysis with formal methods technologies. This paper documents the design of the MorphirIR language and the larger Morphir ecosystem with an emphasis on how they benefit from and enable formal methods for error checking and bug finding. We also report our initial experiences working in this system, our experience using formal validation in it, and identify open issues that we believe are important to the Fintech community and relevant to the research community.