We worked with a fintech startup that was modeling the health of portfolios of consumer loans. The company relied on legacy Excel models to run their analyses, a time consuming and unstandardized process.
The models were built without a view towards consistency, reproducibility, and on-going maintenance. The company wanted to port from Excel into a form that would be scalable, traceable, and production ready.
We ported the models into a programmatic framework with documentation and adhering to sound software development principles. In the process, we also uncovered some areas where the original models were theoretically in conflict and we resolved those inconsistencies.
At the conclusion of our overhaul, we provided a solution that saved time and reduced risk in the analysis process without compromising flexibility.