SUCCESS STORY

CASE: AI-DRIVEN TEST CASE AUTOMATION FOR
HARDWARE-INTEGRATED SYSTEMS 

In a recent collaboration with a leading industrial technology company, we embarked on a project to automate test case generation using AI. The goal was to streamline the process of converting complex, manually written requirements into executable test cases, reducing the manual time and effort required. This effort is part of a broader initiative to optimize testing efficiency with AI, and further our mission to improve the overall quality of software and hardware systems.

PROJECT PHASES

Initial Workshop
Requirement Identification
Follow-Up Workshop
NDA and Secure Collaboration
Proof of Concept (PoC) Development
PoC Evaluation

TESTING

We began the project with a complimentary, exploratory workshop to gain an understanding of the client’s specific challenges and objectives. Together with the client, we identified critical areas for test automation based on their current testing setup and requirements.

A second workshop was conducted to refine our approach, validate initial findings, and agree upon the objectives of a Proof of Concept (PoC). Over an eight-week PoC phase, we collaborated closely with the client to implement AI-enabled test case generation tailored to their needs. This phase involved converting manually written requirements into structured, human-readable Gherkin test cases, supplemented with code generation for test steps.

We established a comprehensive NDA to protect the client’s proprietary information and enable open collaboration. The final stage focused on evaluating the outcomes of the PoC and defining the next steps for future automation efforts.

Challenges we IDENTIFIED

Domain Complexity
Working with hardware-based systems required deep domain expertise, which was sometimes confined to the internal knowledge of subject matter experts, as well as integration with the hardware system under test, signal generation hardware, and monitoring software APIs.
Manual Requirement Management
Despite the desire for automating the whole testing process, the system requirements were still largely managed manually, a practice that is expected to continue for the foreseeable future.

Unstructured Documentation
Converting documentation written for human readers into standardized requirements, test cases in Gherkin, and eventually executable test scripts was a complex task for AI tools and required considerable human oversight.
Next Steps in Automation
The PoC covered automation of test case generation for one specific feature. Next steps include scaling up the PoC to cover all features, and, in parallel, exploring and implementing viable validation strategies for AI-enabled test case generation in general.

KEY BENEFITS FOR THIS PROJECT

For the Client –The client got to explore an area of test automation where there are potentially significant time savings to be gained, and which the client might otherwise not have had the time to do. The first step towards the goal of having testers focus on authoring requirements while AI handles the repetitive aspects of implementation has been taken.
Mutual Learning – Both the client and our team gained valuable insights into the practical applications of AI in test automation, resulting in learning benefits for both parties.
Scalable Solutions – As a provider, this project enabled us to refine our AI-driven automation techniques, which can now be applied to other clients facing similar challenges.
Innovation Leadership – This collaboration positioned both parties as innovators in the field, driving forward the adoption of AI in test automation, and providing our team with confidence in proposing similar joint innovation initiatives with future clients.

Conclusion and RESULT

Modernizing test case creation with AI
Reducing time spent on repetitive tasks
Allowing testers to focus on qualitative work

This project highlights the potential of using AI to modernize test case creation from manually written requirements, thus having testers spend less time on tedious and repetitive tasks and increasing the time they spend on qualitative work. By partnering with clients open to new ways of working and innovation, we continue to increase efficiency in automated testing, to be able to offer better testing strategies and improving the quality of our clients’ software systems.