Full-Cycle Software Development for a BioTech Company

Our Customer

A biotech firm partnered with us to pioneer a transformative approach to biomedical innovation. Their groundbreaking solution focuses on inducing catalytic failure in real-world human models of disease, utilizing organs from deceased donors unsuitable for transplantation. This unconventional approach harnesses the untapped potential of these organs, providing profound insights into human disease and therapeutic responses.

The Obstacles They Faced

The client’s needed quickly assembling a skilled development team to create a solution from scratch, covering the entire software development process. Moreover, they had to transition their successful proof-of-concept into a reliable and scalable technology platform with an MVP launch.

How We Helped

Romexsoft swiftly assembled a dedicated team of experts, successfully ran the project, and built an entire comprehensive software solution from the ground up.

Team Composition

With an aim of meeting all the requirements, we suggested assembling a team of software development experts. The managed team comprises a solution architect, project manager, UI/UX designer, back-end and front-end developers, data engineer, QA engineer and DevOps engineer. This dedicated squad of experts enables us to build the MVP as quickly and efficiently as possible within the established budget.

In order to get there, we conducted 38 interviews in a month before swiftly launching the project. Within four weeks, half of the team had already started performing tasks on the project. Romexsoft assembled and put the entire team to work as soon as in just two months, and what allowed us to perform this so seamlessly and promptly was the strategic approach and our expertise in managing onboarding.

MVP’s Infrastructure

The client’s application infrastructure was built within the cloud-based Amazon Web Services (AWS) environment, which was set up and configured by us from the ground up. We also adhered to the Pillars of the AWS Well-Architected framework, utilizing standard AWS technologies.

Employing Infrastructure as a code (IaaC) approach, specifically the Terraform, allowed us to seamlessly build, modify, version, maintain and replicate the solution’s infrastructure. The application is provisioned in the isolated Virtual Private Cloud (VPC) that offers advanced security features including security groups and network access controls.

For the main execution environment, we configured Amazon ECS on AWS Fargate, eliminating the need for managing clusters of virtual machines to run containers. This approach removes the complexities of choosing server types, deciding on cluster scaling, or optimizing cluster packing.

The initial infrastructure design includes autoscaling, the Application Load Balancer (ALB) for handling load spikes, and the Multi-AZ feature to eliminate hardware and software failures. The client will adopt a microservices architecture to deliver services based on the business domain requirements.

Application Architecture

The adoption of a microservices architecture allows for the easy scaling of individual services, isolating failures to prevent widespread system disruptions. Microservices provide the flexibility to use different programming languages (Python or NodeJS in our case), which simplifies development and deployment. With microservices, we have achieved faster time-to-market by developing and releasing new features independently. This type of infrastructure also proves to foster experimentation and innovation. Enhanced data security is a plus, thanks to the compartmentalization of sensitive information and secure APIs.

The client has opted for Java Spring Boot framework as the primary backend technology stack. This framework provides a quick and efficient environment for Java software development, reducing code complexity, streamlining testing, and enabling easy integration with various databases and services all the while facilitating dependency management and offering embedded server support.

Amazon Managed Streaming for Apache Kafka (MSK) is to be used as the main event-driven communication platform between microservices.

AWS Aurora with read replicas is used as the main database layer, offering a high-performance storage subsystem, fault tolerance, built-in clustering, replication, and autoscaling of read replicas.

AWS ElastiCache is utilized as the caching and user sessions distribution layer, providing enhanced security, reliability, scalability, and performance compared to open-source in-memory storage alternatives.

The front-end layer is implemented as an Angular single-page application (SPA) hosted as a static website on S3. Communication between the front-end and back-end is to be done through the REST API protocol.

Two separate environments, Sandbox and Production, have been selected for development and production use. The environments are isolated in separate AWS accounts under the Client’s AWS Organization. This approach ensures needed account segregation and security compliance. It proves especially useful in case of new product features being delivered for testing by the development team to be sure that production remains stable during the testing activities.

Application deployment is orchestrated using AWS CodePipeline, CodeBuild, and CodeDeploy services. The automated CI/CD process accelerates code delivery, facilitates rapid failure identification through automated testing, and ultimately reduces costs.

MVP of Biomedical Software – Architecture diagram

In general, this project represents the power of partnership and innovation in biotech software development and solving complex challenges and achieving transformative outcomes for the client. The successful launch of the Minimum Viable Product MVP is a critical milestone that helps validate the feasibility and potential of the whole client’s clinical research system. This endeavor commenced by Romexsoft also paved the way for further development.

The successful foundation and launch of the MVP has led to the following achievements:

• Streamlining clinical research
  We have addressed the client’s fundamental business need, which involved the ability to conduct research easily, conveniently, and efficiently, with minimal human factor errors.
• Scalable architecture
  Our experts delivered a robust and scalable platform, ensuring that it can accommodate further growing demands of clinical research. Besides, this architecture is designed to support the integration of additional features as the system evolves.
• Streamlined development
  The usage of the Infrastructure as Code (IaaC) approach with the Terraform has not only optimized the development workflow but also laid the groundwork for the system’s scalability and resilience.
• Enhanced reliability
  With the system’s reliability being a top priority, rigorous testing and quality assurance procedures have been implemented to minimize the risk of system failures during critical research processes.
• Improved security
  By segregating the development and production environments, sensitive data and production resources are shielded from potential security risks originating in the development phase. This ensures that confidential information and critical assets are safeguarded at all times. Moreover, isolating environments in separate AWS accounts decreases the attack surface, limiting the risk of unauthorized access and potential breaches.