Less Than Five Lines of Code
Software engineer and longtime Mathematica user Chad Slaughter uses the Wolfram Language to facilitate interdepartmental communication during software development. While most programming languages are designed to do one thing particularly well, developers like Slaughter often find that the Wolfram Language is more versatile: “With traditional C++, in order to develop a program, it’s going to take several hundred lines of code to do anything interesting. With Mathematica, I can do something interesting in less than five lines of code.”
When he was working at Enova, Slaughter used the Wolfram Language to build Colossus, an analytics engine that provides Enova’s clients in the financial services industry with instantaneous risk and credit analysis. Slaughter’s team was looking for a programming language that would allow them to deploy software changes without involving the entire engineering team in each new change. The Wolfram Language streamlines the process and saves countless hours of development work by communicating more effectively across teams involved in the development process, prototyping and deploying ideas quickly, and avoiding the use of multiple systems to process internal and external data.
In a talk at the 2015 Wolfram Technology Conference, Slaughter’s colleague Vinod Cheriyan explained that streamlining the production process enables Colossus to significantly outperform its predecessor. Colossus can deploy a model to production in just one and a half to two weeks, where its predecessor would typically take one to one and a half months.
Slaughter’s team also used Mathematica to efficiently manage Enova’s large database of XML credit reports. Credit agencies give Enova reports as XMLs with metadata that is packaged as a PDF or Word document. Slaughter’s team replaced a slower procedural approach for merging data with the Wolfram Language’s functional approach, where pattern matching and accelerating rules allowed them to achieve the same result two orders of magnitude faster.
When we talked with Slaughter about why he prefers the Wolfram Language, he mentioned its power both as a programming language and as a computation engine. By using the Wolfram Language, he is able to dramatically streamline his team’s workflow, bringing testing and production into one efficient system.