A leading InsurTech accelerates DevOps, improving deployment productivity by 95%

About the client

The client is the first cloud-native core platform for insurance enabling insurers to build products and experiences for their customers rapidly. The client brings unparalleled speed, flexibility, and choice, giving consumers more control over their business. The client works with global insurers to modernize their technology, thus making it easier for their customers to focus on creating innovative products to grow their business. Founded in 2014, the client is a pioneer of the first core system and provides open APIs, a product-agnostic data model, and innovative capabilities to manage the policy lifecycle.

Business challenge

The client serves one of the world’s largest and most technologically overlooked industry. Thus its mission is to free the insurance industry from the rigid technologies of the past. Adhering to this mission, the client was experiencing accelerated growth and planning to expand into newer markets.

The client desired to deliver a WOW experience to its existing and new customers by providing high-quality technology to insurers globally. However, its current technology platform was highly complex, with numerous features. The sophisticated software had various build environments and development teams.

Furthermore, manual deployments of builds for several microservices resulted in significant deployment time. The client realized the need for an apt DevOps process to deliver new application features at a higher velocity.

Being a forward-thinking company, the client aspired to streamline the code delivery process with the highest levels of quality. The client decided to upgrade its platform to offer an integrated front end that dramatically simplified the experience for online customers and their agents. With more than 37 tenants, apt tenant management with regard to maintenance of customizations and data isolations was also necessary for market readiness.

Trigent solution

The client partnered with Trigent to augment their DevOps capabilities. As a cloud-native solutions provider, Trigent conducted a thorough assessment and understanding of the existing systems as well as processes and developed an end-to-end DevOps strategy. The strategy focused on minor parts of software instead of large chunks to speed up the integration, testing, delivery, and quality.

To speed up the code delivery process and attain a faster time to market, the team implemented three DevOps practices: Continuous Integration (CI), Continuous Delivery (CD), and Continuous Testing (CT).

Trigent leveraged its DevOps capabilities and created a CI/CD build pipeline using Jenkins. Faster builds were created leveraging Maven. The CI/CD pipelines were built in the AWS environment.

CI enabled the client to significantly improve visibility and communication and identify issues/bugs early. CD enabled the client to increase efficiency and cost-effectiveness by 20%. CT helped the team evaluate each layer of the architecture to avoid bottlenecks and execute the correct set of tests at the right phase. Through CT, over 200+ manual test cases were written to cover all the workflows. The negative and edge case scenarios covering the policy dates and premium calculations were carefully designed, and all the complex workflows were thoroughly tested.

Every CI/CD pipeline had a standard Gitflow branching model. Each pipeline step was launched in a separate container image leveraging Docker with an Engineering Change Request (ECR) to maximize fault isolation. Jira was leveraged for end-to-end project management and encompassed details with regard to bugs, environment creations, and other requirements for the microservices.

Furthermore, DynamoBD was leveraged for a consistent environment creation for users and tenants.

Conceptual DevOps Architecture

Agro CD and EKS clusters were used to effectively deploy the latest codes into the production environment. Automated deployment of codes and builds was possible with the apt monitoring of the clusters and infrastructure, and differences between the two were resolved with the help of a DevOps Bot utility. Additionally, the team built a robust framework for resource and site monitoring leveraging Pingdom, whereas Grafana was leveraged to monitor the infrastructure and log analytics for improved operational efficiency.