Developing Custom AI-Powered Software for a Leading Financial Services Provider

Solution

Our collaboration started with working on the Web Crawler application. After the planning stage was complete, we assembled the core team, which included:

• An Engineering Manager
• Full-Stack Developers
• AI/ML Experts
• Quality Assurance Specialist. 

To deal with the main challenges in developing the Web Crawler app, our team decided to leverage our deep expertise in Artificial Intelligence and Machine Learning. So, we prepared a data set for further model training to begin the development process. To solve the challenge of the irrelevant information that contained target keywords, we came up with sophisticated AI/ML (Doc2Vec and Word2Vec) models. These models allow for analyzing a larger context around a keyword, making it possible to cut off most of the irrelevant data. Using advanced AI/ML algorithms has also allowed us to properly process health savings plans with additional reservations (for example, ones that have to do with discounts) that alter these plans’ execution logic. 

Our project team employed AI and ML algorithms to help reduce the amount of data to be processed by the client’s employees weekly as well. In a user-friendly manner, our app shows the sites where no changes in data have taken place since the previous crawling. Besides, it also prioritizes all the retrieved results, indicating the percentage of relevance for each result. While we were working on preparing a custom model, we used AWS Comprehend to score crawled data relevancy. 

In addition to focusing on optimizing the performance and cost-efficiency of the solutions we were tasked to develop, we also:

• Proactively optimized the client-provided data storage architecture, which has resulted in significant cost savings; 
• Our Backend developers implemented the front end of the Web Crawler application, further improving the cost efficiency for the customer. 
• Embraced the modular monolith architecture for our API to deliver a high-quality, scalable, and efficient solution.

Fully satisfied with our performance and speed during the Web Crawler app development, the high-ranking technical officers approached us to develop a Data Collection application. 

The client already had an app developed by another vendor, but it failed to comply with current business demands. Our team made an audit of the existing app and redesigned a major part of its architecture. This allowed us to rule out 90% of the possible data losses and significantly sped up the performance. 

Additionally, we redesigned and dramatically expanded the data annotation functionality of the app and developed from scratch its advanced data validation capability. Our developers put in additional effort to meet the deadline and implement the system GUI in full sync with the client’s requirements. 

After this success, another legacy transformation project for Morningstar, Inc., followed, as we completely overhauled the client’s Kessler App into a full-blown Cloud application. By skillfully leveraging the potential of AWS, we’ve achieved the required high-load capacity and significantly increased the system’s processing speed.

To achieve the required integration with the client’s aggregated accounts system, we interacted directly with several different internal departments to work out the best solutions possible. 

We introduced several business-critical features in the revamped Kessler App, including: 

• Advanced investment analytics functionality with watch lists, charting capability, Investment Calendar, Portfolio Management markets’ comparison, and investment analytics.
• Expanded the platform’s advanced search (Screener) functionality significantly, which enabled searching for stocks with a set of criteria and grouping the found stocks into criteria-based collections.