Building a Data Analytics Pipeline On Cloud Data For Fintech

Client Profile

A FinTech company based out of Dubai in UAE, allows for a shop now experience wherein customers can make purchases without paying the total amount upfront and can pay later (BNPL) in equal installments. The company was founded in 2020 and offered its services in UAE and Saudi Arabia via the app and web-based platform.

Key Requirement

The management team sensed a considerable gap in business and technology wherein they realized that the MVP they developed over time was not yielding results as expected. In addition, the development process was moving slowly and needed to be accelerated. Besides, the core team needed to focus on the business with reduced dependability on improving the application's stability. Given this situation, rewriting the application architecture was not viable as it included a lot of time, effort, and additional costs.

Major Challenges

The team came across multiple roadblocks, including infrastructure hindering the business from progressing quickly. They wanted to move fast with an enhanced IT infrastructure monitoring strategy. And hence, felt the need to create more mature and efficient development and deployment cycles. Some of the significant challenges are listed below: The BNPL market is moving quickly, and the FinTech company needed urgent attention for technical advancement As a growing company, they wanted to rapidly expand its platform capabilities for faster GTM (go-to-market) Limited workforce with necessary new age skill-set Need for a technical partner to understand the current challenges around the platform and provide the best possible solution in a short time

Our Approach

Consulting / Architecture Deep Dive

Flentas did a thorough deep dive of the application and infrastructure architecture through Consulting

The outcome of the above exercise was a detailed report of gaps/improvements in the overall application and Infra landscape

Flentas recommended a solution approach for each gap/improvement discovered during the consulting sessions

Analytics Pipeline

As a startup in the micro-lending space, the team needed detailed and customizable Financial Analysis Reports to understand credit risk and portfolio performance

Flentas helped build a robust Analytics pipeline on AWS using serverless components to keep the cost at a minimum

AWS Services used: S3 (Data Lake), Athena, QuickSight, Lambda, Glue (ETL), DMS

Middleware for Third-Party Integrations

Third-Party integrations are inevitable for any enterprise architecture

Flentas helped architect and build a highly resilient and robust middleware solution on AWS for any third-party integrations for current and future needs

AWS Service used: SQS, K8 Pod-based workers

Observability

Microservice architectures become more and more complex and harder to maintain and monitor as the number of microservices grows

Having a solid observability stack is imperative for any serious business looking to run its operations smoothly

Flentas helped define, architect, and implement the Observability solution for their platform, which provided not only Operational tracing (API performance) but also Business/Contextual tracing (low-level Audit)

Tools used: Zipkin, ElasticSearch, Spring Sleuth

Measuring Success

Scalable Data Analytics Pipeline We assisted in building an analytics baseline within three months and shared detailed reports with the core team. This module is made to scale for a massive amount of data and analysis, which has benefited the business team to access required reports directly and need not depend on core systems.

Enabled free-flowing interactions through Middleware Middleware enabled the free flow of information in various business systems, and it helped the client to seamlessly integrate third-party applications with its core system. This further helped to improve collaboration between multiple functions.

Increased productivity, detection anomaly & visibility We designed an observability solution best fit for the client which helped them gain faster TAT (Turnaround time) for any issues in the application as it allowed for detecting anomalies and better visibility. It eventually helped better the service availability and uptime.