Up to 87% reduction in code conversion time using Generative AI for SpiderRock

Accelerating legacy migration with AI-powered automation and accuracy

SpiderRock Platform is a technology provider based out of Chicago, Illinois, that creates and deploys trading workflows, innovative routing techniques, and risk management solutions. Their clients include large asset managers, proprietary trading firms and trading desks around the world. Their multi-tenant, high-performance cloud-based trading system helps clients source liquidity intelligently and at scale across the global markets.

The client approached us for assistance with a large legacy codebase written in Perl that they were keen to convert to a more maintainable Python based solution. The goal was to identify and implement an innovative conversion solution leveraging automation and AI technologies to reduce time and improve accuracy compared to a manual process.

SpiderRock had a legacy codebase written in Perl, which became difficult to maintain and update. By converting the code to Python, the customer can leverage a more modern and actively supported programming language.

The key challenge of the project was the sheer volume of code that needed to be converted. The codebase consisted of 88 Perl files and over 35,000 lines of code. Manually converting this code would have been costly and time-consuming. If developers are unfamiliar with Perl it can introduce errors too which take time to diagnose and fix.

To solve this problem the team leveraged ChatGPT Plus, which helped accelerate the conversion process and increased overall accuracy. A baseline was established by manually converting Perl files of varying lengths, and five Perl files were then converted using ChatGPT. The translated files were validated by comparing them with the actual output files. The result was up to 87% less conversion time.

The PoV highlighted that the use of ChatGPT can result in significant time savings for the initial conversion process. While this looks promising it is not without limitations. There is a restriction on the length of code that can be processed, which can create bugs that have to manually fixed. For larger lines of code, breaking down the files into smaller parts can yield better outcomes when utilizing ChatGPT. Nonetheless, it remains a valuable resource for developers to improve their Perl/Python skills.

• Conversion efficiency Up to 87% less conversion time than manual processes
• Developer augmentation Generative AI used to accelerate conversion and comprehension
• Maintainability Python’s readability and ecosystem improve long-term maintenance and speed of updates
• Quality assurance Outputs validated against actual files; larger files handled via segmented conversion
• Practical constraints Code-length limitations require chunking and targeted prompts to minimise bugs
• Scaling AI innovation We provided a plan to operationalise generative AI in the conversion workflow