Blockchain

IBM + AWS: Transforming Software Development Lifecycle (SDLC) with generative AI

Generative AI is not only changing the way applications are built, but the way they are envisioned, designed, tested, documented, and deployed. It’s also revolutionizing the software development lifecycle (SDLC).

IBM and AWS are infusing Amazon Bedrock generative AI capabilities into the IBM® SDLC solution to drive increased efficiency, speed, quality and value in every application lifecycle consistently and at scale. And

The evolution of the SDLC landscape

The software development lifecycle has undergone several silent revolutions in recent decades. The dawn of software application delivery started with large backend programs running on specialized servers, and the advent of PCs introduced the desktop into the application development lifecycle mix. Later, the rise of the web and social media paved the way for increased speed and scale in the number of applications and their interconnections. The appearance of mobile apps and the cloud incorporated new levels of automation and agility into the development lifecycle. As a result, the SDLC for application delivery became radically different from what it used to be. With the advent of gen AI, we are witnessing another transformation in the way applications are developed and delivered.

Figure 1: Application development across decades

The rise of generative AI in SDLC

Adoption of generative AI in the end-to-end SDLC brings numerous benefits, such as accelerating development time, improving code quality and reducing costs. By using generative AI, we can reduce the time-to-market for our clients. It also improves the effectiveness and consistency across tasks and participants by reducing the number of handovers, automating or removing low-value mundane tasks, and facilitating access to knowledge and onboarding.

Generative AI drives these benefits for the complete end-to-end application lifecycle across all stages and procedures for all participants from ideation to deployment.

These are the key SDLC areas gen AI can make an impact:

Figure 2: Typical participants through stages and phases of SDLC
  • Business and product owners can use generative AI during ideation and concept generation, requirements identification, prioritization and planning, as well as other activities such as understanding feedback from users or clients.
  • Analysts and designers can use generative AI to accelerate the creation of prototypes along with detailed functional designs and solution blueprints.
  • Developers and testers can use generative AI to define (and reuse) solution architectures, create and reuse technical designs, create code and logic. They can also build (and run) highly automated tests and perform quality and validation procedures. Engineers can use generative AI to activate the application underlying technical environment, on cloud or on premises. They can also perform the promotion and deployment of the application across the different environments and governance gates. IT support administrators and operators can use generative AI across the multiple activities they perform regularly, including monitoring, operation and remediation, incident management including triage and resolution, and service request fulfillment.

Adoption of generative AI in the end-to-end application SDLC brings numerous benefits, such as accelerating development time, improving code quality and reducing costs. This approach not only accelerates individual tasks, but also enables us to perform activities earlier than is possible today, such as validation with business and users. It improves the effectiveness and consistency across tasks and participants by reducing the number of handovers, automating or removing low-value mundane tasks, and facilitating access to knowledge and onboarding.

The new gen AI-powered SDLC solution

Together, IBM and AWS have developed a joint gen AI-based SDLC solution, which is now available on AWS Marketplace. The solution automates the use of company architecture standards, assets, security, available APIs, quality standards and documentation models helping ensure that all artifacts comply with approved and defined policies within the organization’s SLDC.

To achieve the benefits mentioned earlier at scale in a sustainable manner, we apply a thoughtful and deliberate approach for integrating generative AI into every SDLC. Such an approach involves adapting our solution to the reality of each organization’s needs and SDLC, which is essential for achieving optimal results.

An interesting observation and challenge of generative AI is that it often produces different results when given the same input. Just like when two different developers being asked to solve the same problem, generative AI produces similar, but not identical, results. Hence, there’s a need for a new set of frictionless guidelines, guardrails, and controls to achieve quality and consistency with different generative AI results, at scale.

These redesigned standardized procedures are key to delivering high-quality standards throughout, while facilitating handovers between teams, so all team members can understand and work with the results generated by generative AI. Furthermore, the technology can drive increased visibility into which stage of the development process is at, improving project management and tracking.

Our capabilities

It is important to note that standardization and consistency in the SDLC are not achieved solely through generative AI. Our achievement is due to the extensive work done at IBM Consulting®, carefully designing generative AI-based procedures applied across the end-to-end SDLC. We have been adapting and refining our solution for each SDLC stage and task, which allows generative AI to produce consistent and high-quality results. This experience has enabled us to create guided, frictionless procedures adapted to the specific needs of each client to properly address the reality of their SDLC and software landscape.

Solution benefits

Based on data collected from customers that already use this solution, customers can expect to achieve significant benefits and outcomes, including the following:

  • Accelerated development time: Up to 30% reduction in development time, enabling customers to bring their products and services to market faster.
  • Test generation time: Up to 25% of time improvement in unit test generation and test plan scenarios.
  • Improved code quality: Up to 25% improvement in code quality, resulting in fewer errors, reduced rework and lower maintenance costs.
  • Reduction of analysis phase time: Up to 60% reduction in the analysis phase including functional and technical requirements, reverse engineering and documentation enrichment.
  • Enhanced collaboration: Improved collaboration and handovers between teams, enabling customers to work more efficiently and effectively.
Figure 3 gen AI SDLC Solution and existing client’s platform

Our generative AI-powered solution offers several key advantages for developers and companies looking to improve their SDLC:

  • The solution adapts to clients’ needs, including custom components or by using their own reusable front-end components or backend libraries.
  • It is compatible with existing DevOps solutions, like CI/CD tools to start compilation after code generation or kanban boards for obtaining user stories to create a detailed design of the software. This allows clients to quickly integrate it into their current processes. The solution offers greater speed and efficiency in software development, which can reduce costs and improve the quality of the final product. By using our solution, developers can improve the effectiveness of their SDLC, reducing the time and effort required to develop high-quality applications.

Users have the flexibility to accept solution suggestions as-is, ask for an automated suggestion rephrase or make manual modifications to adapt suggestions to their specific needs. During the coding phase, we provide requirements, test cases for Test Driven Development (TDD) and other information to Anthropic Claude Sonnet, and the model will generate the needed code. The developer can modify the code, refactoring it to have the wanted code structure. After all, users are experts in their domain, and the IBM solution provides them with the opportunity to refine and perfect the suggested results according to their specific needs.

Technical architecture

The generative AI-powered SDLC solution uses Amazon Bedrock to consume large language models (LLMs) such as Anthropic’s Claude family of models, and Amazon S3 stores our refined and tailored procedures and prompt templates.

A complete user interface and integration layers are built as containers that run on Amazon Elastic Kubernetes Service (Amazon EKS). IBM uses AWS services to fulfill specific needs. These include AWS Web Application Firewall to secure application endpoints, AWS Application Load Balancer to manage application traffic and Amazon API Gateway to enable connections with external services like CI/CD.  

Security and compliance are key concerns for any business. AWS Key Management Service (AWS KMS) manages keys and the encryption of customer data, enabling customers to adhere to their standards of privacy, security and compliance. External keys can be used depending on where the customer prefers to keep cryptographic material. Integration with AWS CloudHSM or a third party HSM is also possible.

AWS CloudTrail is used to enable detailed audit trails of user and system actions, critical for supporting regulatory audits and helping to demonstrate compliance. AWS Identity and Access Management (IAM) is used to implement granular control over access to data and resources with support for multi-factor authentication. AWS Certificate Manager provide secure management of X.509 certificates for SSL/TLS connections, securing data in transit. AWS Secrets Manager centralizes and secures secrets, such as API keys and data repository credentials.

The preceding architecture offers a robust and secure solution intended to be deployed and integrated at scale by your organization’s unique landscape of technologies and vendors. This includes integration with existing DevOps solutions, allowing you to quickly integrate it into your own current processes.

Exploring generative AI for SDLC?

Our generative AI-infused SDLC solution (available on AWS Marketplace), is revolutionizing the software development lifecycle by providing a faster, more efficient, consistent and secure way to develop software. It combines the power of AWS generative AI technologies with the flexibility and scalability of the AWS Cloud. This enables developers and companies to create high-quality applications in a shorter time frame and at a lower cost.

As an AWS Premier Tier Partner, IBM brings specialized industry and domain expertise to help organizations move beyond pilots and achieve tangible business outcomes with generative AI. With the new AWS GenAI Competency, IBM Consulting has now has 21 AWS competencies and 17 service delivery designations (SDDs). This demonstrates our commitment to helping clients unlock the full potential of cloud-based generative AI solutions.

We are thrilled to see the impact our solution can have on transforming the software development landscape. We’re dedicated to ongoing innovation and improvement to make sure our technology meets the evolving needs of our clients.

Read more about AWS Consulting Services

Was this article helpful?

YesNo


Source link

Related Articles

Back to top button