July 2, 2021

Softensity Teams With RelationalAI on Groundbreaking Tax Analysis App Rewrite for Big Four Accounting Firm


When a Big Four accounting firm turned to RelationalAI to adopt a relational knowledge graph model for their complex tax analysis application, Softensity provided the additional resources necessary to achieve a digital transformation of this scale and scope.


The accounting firm’s tax analysis app was straining to handle large, complex workloads. The app analyzes clients’ securities portfolios to create a general ledger that complies with GAAP and applicable tax laws. Due to the interrelated nature of tax laws, which are not independent of one another, every accounting decision influences tax adjustments that affect other accounting decisions. The complex analysis algorithm needed to account for the intricacies of these various interrelated tax laws, and have the ability to scale in order to accommodate trades that involve billions of shares.

The legacy app performed analysis for each tax law in isolation, which made the process highly inefficient. Implemented in C#, the legacy system was built with hundreds of thousands of lines of code. The siloed analysis functions lacked cohesion, which made these complex calculations take an extensive amount of time to complete. It also made it difficult to scale the data, and to update the app for every tax regulation change.

The Solution

The accounting firm turned to RelationalAI to replatform the app onto a Relational Knowledge Graph Management System (RKGMS) in order to make the app faster, more cohesive, more consistent and easier to scale. While RelationalAI had the expertise to spearhead the project, they relied on Softensity to provide additional resources in order to complete the ambitious overhaul.

Softensity worked hand in hand with RelationalAI and the accounting firm to rewrite the code in order to achieve a much smaller footprint, and a more agile application. Softensity’s team included a Solutions Architect with a high level of expertise in relational databases and knowledge graphs, along with Project Managers, Data Engineers and Developers. Together, RelationalAI and Softensity centralized the business model, incorporated advanced reasoning and implemented relational architecture that made the data easier to scale.

The Results

With the help of RelationalAI and Softensity, the accounting firm became the first of the Big Four to adopt an advanced platform of this scale based on microservices architecture. Moving the app to a RKGMS had numerous benefits. The streamlined codebase is 50 times smaller and requires significantly less maintenance, which reduces both risk and expenses. The app is 10 times faster, and is easier and safer to update without issue. Tax decisions conform even closer to the tax code, and are more consistent and explainable, which reduces risk in the case of an audit. The model is built to ensure data quality and integrity, which is imperative for tax analysis.

Softensity’s initial engagement spanned more than a year, and has since grown and evolved. Softensity’s work with RelationalID and the accounting firm continues, along with a direct engagement to provide the accounting firm’s Tax Services Department with additional resources. RelationalAI continues to build the world’s fastest, most scalable, most expressive and most open knowledge graph management system for a growing number of influential clients.

Watch an interview with RelationalAI CEO, Molham Aref.


Let’s Talk