Transforming Cloud Quality Engineering with Infosys Quality Engineering AI platform

Infosys

By Arvind Sundarraman, Practice Engagement Manager – Infosys
Rahul Deshmukh, Solution Architect – Infosys
Ashutosh Dubey, Solution Architect – AWS
Ashutosh Pateriya, Solution Architect – AWS

As technology rapidly evolves, businesses need to adopt cloud computing to remain competitive. Migrating to the Amazon Web Services (AWS Cloud) can help organizations achieve greater scalability, flexibility, and efficiency. A key part of any cloud migration is implementing the quality control processes and compliance checks. This ensures your cloud environment is secure and meets all relevant regulations.

Testing and verifying new cloud infrastructure and services can be a significant challenge during cloud migrations. Infosys, an AWS Premier Consulting Partner, has extensive experience in cloud migrations and testing. We have developed an AI-powered automated quality engineering solution to address the need for comprehensive quality control, while ensuring the agility and reliability throughout the cloud migration journey.

This blog post delves into how Infosys Quality Engineering AI platform leverages AWS Generative AI services to revolutionize Quality Engineering practices. By harnessing the power of Gen AI, Infosys’ Generative AI Test Platform empowers organizations to address the complexities involved in cloud migration and ensure successful deployments.

Challenges faced by customers during their cloud migration journey:

Traditional approaches to cloud migration testing often lack the agility and efficiency required for digital transformation. Manual testing is prone to human error, leading to delays and increased costs.
Testing often requires using real data, which raises security and privacy risks.
Legacy applications may not have sufficient documentation and test inventory, making it challenging to thoroughly test these applications after migration.
Scripting tools and languages might differ between legacy and cloud environments, making script migration for automated testing cumbersome.
Traditional testing methods may not adequately stress test cloud infrastructure to identify performance bottlenecks under heavy load.
Many existing AI technologies often fail to meet the diverse needs and intricacies of cloud migration testing.

The convergence of AWS Gen AI services and Infosys Quality Engineering AI platform transformation marks a significant leap in modern testing capabilities. By leveraging AWS Gen AI capabilities, the entire lifecycle of testing workflow has been automated, encompassing workload discovery, planning, testcase development, execution, and root cause analysis, which otherwise would have been time-consuming and costly.

Infosys Quality Engineering AI platform overview:

The Infosys Quality Engineering AI platform hosts Gen AI use cases for the migration life cycle, with ready to use plugins for various testing phases as shown below:

Figure1. Quality Engineering (QE) evolution with the advent of AWS Gen AI Services

The diagram represents a structured approach to identifying, planning, developing, and executing tests, setting up the necessary environments, and managing defects through automated and manual processes.

It includes five key stages:

Discovery & Risk Assessment (requirements): This stage involves analyzing workload patterns, security needs, and cost considerations to ensure an optimized cloud migration test strategy.
Test Planning: Define approach for testing which includes test data generation, test case optimization & prioritization.
Test Case Development: Here, specific test cases & automated scripts are generated based on the test plan.
Environment Setup: This stage involves automated validation of the testing environment and test data generation.
Defect and Root-Cause Analysis: Defects are classified & prioritized, with self-healing for simple defects.

Solution Architecture:

The Infosys Quality Engineering AI platform solution makes use of fully managed services by AWS. The solution uses the Amazon Bedrock foundation models (FM) which references user knowledge base by deploying Retrieval Augmented Generation (RAG) technique to improve the accuracy of results for different use cases with advanced prompting techniques.

Figure2. Infosys Quality Engineering AI platform AWS architecture

This diagram illustrates a workflow for leveraging AWS services for implementing Gen AI use cases to accelerate the cloud transformation journey as following:

User Authentication: Users authenticate and log into Infosys Quality Engineering AI platform through url hosted on Amazon Elastic Kubernetes Services (EKS) services and creates apps needed for different Gen AI use cases.
Create Knowledge base: Authenticated users upload the documents such as build specifications, architecture and design documents, legacy code and test cases to build knowledge base needed for model learning, fine-tuning or training. These documents are then stored on storage for specific Generative AI use cases:
- Amazon Simple storage service (S3) bucket for knowledge base storage.
- Amazon DynamoDB for key-value and document data storage.
Model Customization and Training:
- Prompting techniques like zero-shot or one-shot or few-shots, chain-of-thought, role-play, iterative, Context, etc.
- RAG (Retrieval-Augmented Generation) techniques – knowledge data is fed into an Amazon Bedrock Embedding Model.
- Model fine tuning for specific task or domain by training relevant dataset using Amazon Bedrock Custom Model Training Job. It further processes and presents custom models using foundation models like Amazon Titan, Claude 3, Llama 2.
- Workflows using Bedrock and flowise to create workflow using agents, Langchains, utilities, connectors, etc.
Infosys Responsible AI: Implement the responsible AI ensuring privacy, protection from toxicity, fairness, transparency, security, and safety for organizations by continuously tracking and monitoring of the metrics.
User execute apps on Infosys Quality Engineering AI platform: Authenticated users interact with Infosys Quality Engineering AI platform through a user interface (UI) supporting workflows, agents, Large Language Models (LLMs), and other capabilities for Gen AI use cases via inputs like prompt, user stories, etc. needed for the trained Custom Bedrock models. User then executes the application to generate the outputs like Test cases, automation scripts (Python, Pester-PowerShell, Unified Functional Testing (UFT) Scripts, SQL, Selenium, Java, etc.), and defect analysis reports.
Storage of Data: Training, validation, and output data are stored in another Amazon S3 Bucket and Amazon DynamoDB.
Testing and Defect Management: Finally, the generated output (scripts, test cases, etc.) are used for executing testing and defect management.

Infosys Quality Engineering AI platform use cases:

Infosys Quality Engineering AI platform focuses on creating custom solutions using prompt engineering and specific training for LLMs to address quality engineering challenges. Built on a connected ecosystem, it enables collaboration with stakeholders and, with an advanced toolkit, streamlines and enhances the entire testing process.

The platform hosts several pre-tuned and customized use cases specifically to aid in accelerating the quality engineering life cycle, as described below:

Automated Test Case Generation: Gen AI quickly creates diverse, thorough test cases for cloud migration by analyzing application ecosystems and usage patterns.
Synthetic Data Generation: Gen AI generates synthetic datasets that mimic real-world data, ensuring comprehensive testing without using sensitive production data.
Cross-Platform Script Migration: Gen AI transforms automation scripts to be compatible with cloud-based platforms, speeding up cloud-based testing transitions.
Performance Testing Optimization: Gen AI enhances performance testing by simulating realistic workloads, identifying performance bottlenecks and scalability issues.
Security Testing Automation: Gen AI automates security testing by simulating cyber threats and vulnerabilities, strengthening cloud migration security.
Anomaly Detection using VAEs: Variational Autoencoders detect anomalies in system performance during cloud migration by analyzing data reconstruction errors.
Dynamic Test Environment Simulation: Reinforcement Learning dynamically adjusts test parameters in real-time based on ongoing tests and system feedback.
Script Generation for Automated Testing: Code generation models create and update automated test scripts based on user inputs and historical data, reducing manual maintenance.
Generate Test Scenarios from Application Code: Gen AI analyzes application code to understand functionalities and generate comprehensive test scenarios covering various execution paths.

Infosys Quality Engineering AI platform benefits:

Infosys Quality Engineering AI platform offers significant advantages for Quality Engineering teams, leading to improved efficiency, effectiveness, and overall software quality.

Automated Test Case Generation: Increase test coverage by 20-40% or more compared to manual test case creation by fine-tuning foundation models with user knowledge base.
Improved Test Data Management: Reduce manual test data creation effort by 50% or more using synthetic test data generation simulated to real production scenarios using RAG techniques
Early Defect Detection: Increase defect detection rate during unit and integration testing by 10-15% by generating broader range of test scenarios and data.
Regression Test Optimization: Reduce regression testing execution time by 20-40% by focusing on the most critical test cases using risk-based approaches.
Script generation for automated testing: Improves productivity across testing types by generating the automated scripts in range of 20-40%.
Reduced Time to Market shortens the testing cycle by 15% to 20%.
Enhanced Accuracy: Fine-tuned and optimized Gen AI solution consistently achieves 98% accuracy in test execution.
Resource Optimization: Infosys Quality Engineering AI platform can free up QE resources by automating 40% to 60% of tasks.

Summary:

The convergence of AWS Gen AI and Infosys Quality Engineering AI platform represents a quantum leap in modern enterprise validation testing capabilities. By exploring these use cases, it becomes evident that Infosys’ Gen AI Test Platform not only addresses the immediate challenges of cloud migration testing but also fosters a culture of continuous improvement and adaptability within QE practices.

The journey towards cloud-native architectures is an ongoing evolution. If you’re considering migrating to the AWS Cloud and looking for a testing partner, contact us for more information.

.

.

Infosys – AWS Partner Spotlight

Infosys is an AWS Premier Consulting Partner that helps enterprises transform through strategic consulting, operational leadership, and co-creation of solutions in mobility, sustainability, big data, and cloud computing.

Contact Infosys | Partner Overview

*Already worked with Infosys? Rate the Partner

*To review an AWS Partner, you must be a customer that has worked with them directly on a project.