Top 6 innovations from the IBM – AWS GenAI Hackathon

Generative AI innovations can transform industries. Eight client teams collaborated with IBM® and AWS this spring to develop generative AI prototypes to address real-world business challenges in the public sector, financial services, energy, healthcare and other industries. Over the course of several weeks, cross-functional teams comprising client teams, IBM and AWS representatives worked to design, develop and iterate on prototypes that push the boundaries of what's possible with generative AI.

IBM used design thinking and user-centric approach to guide the teams throughout the hackathon. AWS provided enablement sessions and hands-on workshops, providing participants with the necessary knowledge and skills to use AWS generative AI services such as Amazon Bedrock and Amazon Q effectively. The upfront enablement helped teams understand AWS technologies, then put that understanding into practice. Their results will influence the next generation of business solutions that enhance customer experience, boost employee productivity and optimize business processes.

Use case 1: Generative AI for Change Management

A leading financial services organization faced challenges in managing the high volume of change management tickets and identifying potential risks associated with deploying changes to production environments. The team developed a "Generative AI for Change Management" solution to enhance the quality of change management tickets and detect the likelihood of risk and failure based on similar historical data.

This solution used AWS services such as Amazon Bedrock, AWS ECS and Amazon Aurora to build an interactive AI interface. The interface allows change managers to create high-quality tickets and understand potential risks and improvement areas. The solution aimed to mitigate the likelihood of incidents occurring due to changes made in production, improve the overall change management process, and shift focus from ticket creation to quality enhancements.

Use case 2: Intelligent Feedback Analysis

An energy company needed to better understand customer satisfaction and identify areas for improvement based on customer feedback. They created an Intelligent Feedback Analysis tool that automates the extraction and analysis of customer comments and reviews across the energy sector.

Using AWS services like Amazon Q for Business, Amazon SageMaker, Amazon Bedrock and Amazon QuickSight, they used generative AI to identify market trends and analyze feedback sentiment. The AI was also used to classify topics and detect potential bugs for existing features or new feature requests from customer feedback.

The solution provided valuable insights into the company's performance, key trends across the sector, and a comparison with competitors, enabling them to rapidly identify high-value areas for improvement. The data would be accessible to stakeholders through a virtual assistant interface and an accompanying dashboard tool providing valuable insights into the company's performance.

Use case 3: Resilience by Design Advisor

A multinational bank faced challenges in maintaining operational resilience due to complex technology landscapes and regulatory scrutiny. They developed a "Resilience by Design Advisor" to address these challenges.

This solution used AWS services such as Amazon Bedrock, Amazon ECS and Amazon S3 to assess solution design documents, stay updated with regulatory updates and industry best practices. It also incorporated the bank's technology resilience framework. The Resilience by Design Advisor enhanced the bank's ability to identify and implement resilience measures in their applications, helping ensure compliance with regulations and maintaining high availability for customer services.

Use case 4: Citizen Feedback Analysis

A government agency aimed to derive actionable insights from citizen feedback provided through their feedback service system. They used generative AI to develop a solution that might effectively analyze and extract valuable information from unstructured feedback data.

By using AWS services such as Amazon Comprehend, Amazon Bedrock, Amazon Aurora, Amazon DynamoDB, the solution can process text feedback, redact personally identifiable information (PII). It also identifies key topics and sentiments, and generate actionable insights to improve the service system.

Use case 5: Generative AI powered Clinical Coding Assistant

A healthcare organization sought to streamline the clinical coding process for electronic patient records. They developed a 'Clinical Coding Assistant' solution that used natural language processing (NLP) and generative AI to extract and convert medical notes into standardized codes.

Using AWS services like Amazon Bedrock and Amazon Aurora, the solution might accurately process and code medical documentation, reducing the time and effort required for manual coding. This might result in an annual saving of GBP 700,000, which might fund 20 extra nurses.

Use case 6: Self-healing CI Pipeline

A government agency faced challenges in maintaining an efficient and reliable continuous integration and deployment (CI/CD) pipeline. Manual intervention was required to diagnose and fix pipeline issues, leading to delays, increased workload for engineering teams, and potential downtime that might impact product releases. Also, critical information for solving pipeline problems was scattered across various documentation sources, making it difficult to access accurate and up-to-date information when needed. To address these challenges, the organization developed a "Self-Healing CI Pipeline" solution.

Using AWS services such as AWS Distro for Open Telemetry, AWS X-Ray, Amazon CloudWatch, Amazon DevOps Guru and Amazon Bedrock, the solution aimed to automatically detect and resolve CI/CD pipeline failures. When a build or deployment pipeline failed, the solution was designed to receive the failure logs and process them. It would pinpoint the root cause of the issue. Then, it would proceed to either fix the identified problem automatically and rerun the pipeline, or it would alternatively provide detailed explanations accompanied with suggested remediation actions.

The goal of this approach was to enhance efficiency in resolving errors, decrease downtime, and boost the overall dependability of the CI/CD pipelines. This can result in quicker product releases and allow engineering resources to concentrate more on enhancing the organization's AWS estate.

IBM and AWS: Unlocking innovation

The gen AI hackathon fostered innovation, collaboration and the development of groundbreaking solutions. Participants gained valuable insights into the potential of gen AI technologies and how they can be used to drive digital transformation and operational excellence. "We'd been talking about this for over a year," said a client about their project, "and now we've got it done in just eight weeks!"

IBM, an AWS and Premier Tier Partner holding the AWS Generative AI Competency, and AWS jointly provide a platform for client teams to design and prototype innovative solutions that use the latest AI technologies. The AWS Generative AI Competency differentiates IBM Consulting® as an AWS Partner that has demonstrated technical proficiency and proven customer success in helping enterprises operationalize and derive value from AWS generative AI technology. IBM and AWS are committed to lowering the barriers to AI experimentation by providing comprehensive support for pilot projects, including infrastructure credits.