Share

cover art for Evo DK #490 - Is AI Improving Software Quality, or Is It Just Speeding Up the Process?

Evolution Exchange Denmark Podcast

Evo DK #490 - Is AI Improving Software Quality, or Is It Just Speeding Up the Process?

•

Today's episode is hosted by George Clark and features Chituru Chinwah, Senior QA Engineer at Corti, Roy Alexander Boven, Senior Test Manager at The Tech Collective, and Rikita Holkar, Global Test Management Specialist at AGC Biologics.


The conversation explores whether AI is genuinely improving software quality or primarily accelerating development. The speakers examine how AI is changing the relationship between development and testing, highlighting why faster code generation does not necessarily mean faster or more effective verification. They discuss the importance of risk-based testing, human accountability and clear quality standards, particularly in regulated industries where reliability, compliance and proper validation are essential. The discussion also considers how organisations can communicate these challenges to leadership while balancing delivery speed with confidence in the final product.


The conversation further examines how QA teams can adapt to AI-driven development without becoming bottlenecks. Key themes include shifting testing earlier in the development process, improving requirements and business alignment, using AI for code reviews and test generation, and strengthening observability to identify issues in production. The speakers also emphasise the continued importance of human judgement, technical expertise, exploratory testing and quality awareness across development teams. Ultimately, AI can help teams work faster and support better quality, but meaningful improvements still depend on effective testing, appropriate oversight and clear accountability.

More episodes

View all episodes

  • Evo DK #495 - What Are the Biggest Impacts We Can Expect to See from AI in the Next 5 Years?

    48:52|
    Today’s episode is hosted by Alex Robinson and he is joined on the podcast by Claus Grand Bang, Head of Data & IT at DanBred P/S, Rasmus Sten Andersen, Associate Director, Product and AI at Novo Nordisk A/S, and Stefi Savova, Senior Manager, Digital Twins & Data Analytics at Danfoss A/S.The conversation explores the biggest impacts AI could have on people and the workplace over the next five years. The speakers discuss how organisations can encourage AI adoption, the importance of change management, and why introducing new technology requires more than simply providing employees with the right tools. They also examine how AI is reshaping professional responsibilities, blending traditional roles and creating new ways for people and AI agents to work together.The discussion looks beyond productivity to consider the human experience of work, including whether automation will remove tasks people genuinely enjoy, how much control humans should retain over AI-driven decisions, and what meaningful human oversight looks like as AI agents become more autonomous. The speakers highlight the importance of building strong data foundations, applying AI where it creates genuine business value, and keeping adoption human-centred. Ultimately, they consider how organisations can use the efficiency AI creates while ensuring the work that remains is still engaging, valuable and rewarding.
  • Evo DK #494 - Has Agile Improved Software Quality?

    01:26:39|
    Today’s episode is hosted by George Clark and he is joined on the podcast by Nicolai Nielsen, QA Manager at Sweco Danmark A/S, George Zubrienko, Lead Software Engineer, Data & AI Platform at ECCO, Mike Ager, Scrum Master at Grundfos and Andreas Bastian Hansen, Senior Test Manager at Demant A/S.The conversation explores whether agile has genuinely improved software quality, examining the challenges organisations face when adopting agile practices and the importance of recognising the time, investment and organisational changes required to make the transition successful. The speakers discuss how agile needs to be adapted to different teams and environments, the value of team ownership, and the importance of maintaining direct collaboration between developers, stakeholders and end users.The discussion also looks at why software quality can suffer when testing is treated as a final step rather than an ongoing part of development. Key themes include early testing, clear requirements, reducing technical debt, encouraging cross-functional collaboration and involving users in feedback and release processes. The speakers ultimately agree that agile can improve software quality, but only when organisations prioritise communication, continuous improvement and quality alongside delivery speed.
  • Evo DK #493 - Why Do Software Projects Still Struggle with Quality?

    01:04:14|
    Today’s episode is hosted by George Clark and they are joined on the podcast by Denitsa Zhekova, QA Architect at Merkle Danmark, Sayma Nasrin, Senior Software Quality Assurance Engineer at Hyperbolic ApS, Haiper Safi, Cross Functional Test Manager at NNIT A/S and Max Møller Viuff, Senior QA Specialist at Lemvigh-Müller.The conversation explores why software projects continue to struggle with quality and how artificial intelligence is reshaping software testing and quality assurance. It considers the opportunities and challenges of AI-driven testing, including automated test generation, agentic AI, potential hallucinations and the importance of maintaining human oversight, critical thinking and domain knowledge throughout the development process.The discussion also examines why software quality is often compromised under deadline pressure and whether faster development cycles are genuinely improving software reliability. Key themes include balancing speed with quality, strengthening QA practices, involving testers earlier in the development lifecycle, maintaining clear requirements and documentation, and establishing responsible AI adoption through effective collaboration, continuous learning and human judgement.
  • Evo DK #492 - What Separates Good Testers from Great Testers?

    55:17|
    Today's episode is hosted by George Clark and features Maurizio Piccini, Senior Technical Product Owner at VELUX, Shahid Nadeem, Software Test Manager at LINAK, Alejandra Ugarte Bedwell, Senior Quality Assurance Engineer at Public, and Frederik Skov, Head of Quality and Test at KOMBIT.The conversation explores what separates good testers from great testers, highlighting the importance of taking ownership, demonstrating curiosity and looking beyond assigned testing tasks. The speakers discuss how great testers challenge requirements, understand business objectives, consider the end user's perspective and identify potential issues before development is complete. They also emphasise the importance of involving testers early in the development process through a shift-left approach, allowing them to contribute to requirements reviews, collaborate with developers and help shape product quality from the beginning.The discussion further examines the role of soft skills in building effective quality assurance teams, including communication, trust, adaptability and the ability to challenge assumptions constructively. The speakers share their experiences working in multicultural environments and explain how different perspectives and communication styles can strengthen collaboration. They also consider how AI is changing the future of software testing, with human judgment, empathy, critical thinking, ethical awareness and business understanding becoming increasingly valuable. Ultimately, the conversation suggests that great testers will evolve beyond test execution into quality strategists who use AI as a supporting tool while maintaining responsibility for delivering products that genuinely meet users' needs.
  • Evo DK #491 - AI in Testing: Revolution or Hype?

    50:32|
    Today's episode is hosted by George Clark and features Henrique Fernandez, Global IT Release Manager at AGC Biologics, Preeti Mishra, Senior Quality Assurance Engineer at Inpay A/S, Sayma Nasrin, Senior Software Quality Assurance Engineer at Hyperbolic ApS, and Kenneth Kim Knudsen, QA Test Manager at IPW Systems A/S.The conversation explores whether AI represents a genuine revolution in software testing or is surrounded by more hype than practical value. The speakers discuss how AI can accelerate test case generation, automation, documentation and test data creation, while highlighting the importance of human oversight to ensure testing remains aligned with business requirements and actual risks. They also examine the limitations of AI-generated tests, including duplicated scenarios, incorrect assumptions and false confidence from large numbers of passing tests. The discussion emphasises that test quality should be measured by its ability to identify meaningful defects and improve product reliability, rather than simply the volume of tests produced.The discussion further examines how AI is changing the role of QA professionals, creating opportunities to focus more on exploratory, performance and security testing, as well as product knowledge and early involvement in development. Key themes include measuring AI's return on investment through meaningful quality indicators, strengthening collaboration between developers and testers, and ensuring that new features deliver genuine value to customers. The speakers agree that AI can improve productivity and support better testing practices, but its effectiveness ultimately depends on the expertise of the people using it, clear requirements and sound quality processes.
  • Evo DK #489 - Who Owns Quality When Everybody Owns Testing?

    01:02:24|
    Today's episode is hosted by George Clark and features Tommy Mogensen, Test Automation SME & Manager at NNIT A/S, Augustin Gottlieb, Director of QA & Automation at Merkle Danmark, Thomas Host Andersen, Engineering & Technology Leader, and Linus Vogel, Advanced Test Manager at DSB København.The conversation explores the question of who owns quality when everyone is responsible for testing, examining the importance of clear accountability, shared responsibility and decision-making within software development teams. It considers how organisations can establish effective quality ownership while ensuring that developers, testers and business stakeholders work together to maintain high standards. The discussion also highlights the importance of risk assessment, traceability and transparent communication when making decisions about product releases and accepted defects.The discussion further examines how AI and agentic AI workflows are transforming quality assurance and test automation, creating opportunities to improve productivity while introducing new challenges around reliability, alignment and governance. Key themes include human oversight, AI-driven testing, quality risk management, appropriate guardrails and the importance of maintaining strong testing practices as development processes accelerate.
  • Evo DK #488 - What Actually Makes a Great Test Engineer?

    57:01|
    Today’s episode is hosted by George Clark and features Bogdan Stoenescu, Engineering Manager, QA at Merkle Danmark, Akhilesh Sinha, Quality Engineering Manager at Pandora A/S, Claus Petersen, Test Manager at Q Nation Aarhus A/S, and Søren Thomsen, Senior Manager, Production Test Equipment at RTX A/S.In this episode, the panel explores what actually makes a great test engineer, discussing the professional skills, mindset, and experience that help testers deliver meaningful improvements to product quality. The conversation highlights the importance of curiosity, analytical thinking, understanding how systems and businesses operate, and developing the intuition to anticipate potential failures. The speakers also discuss how these capabilities can be cultivated through hands-on experience, product knowledge, imagination, and a willingness to challenge assumptions.The discussion also challenges common software testing myths, including the idea that quality should be addressed only at the end of the development process, that a tester’s sole responsibility is to find bugs, and that everything should be automated. The panel emphasises that quality is a shared responsibility throughout the product development lifecycle, supported by early collaboration, clear testing strategies, and a balance between automation and exploratory testing. They also explore how QA engineers and developers can share testing responsibilities through open communication, defined ownership, and an understanding of each project's needs.
  • Evo DK #487 - How AI Can Break Down Barriers Between Tech Teams and Business Teams

    01:01:38|
    Today’s episode is hosted by Louis Bradley and they are joined on the podcast by Andreas Steinbacher, Head of Software at Novo Nordisk A/S, Rasmus Kimmer Werther, Group Head of Data & Analytics at Colosseum Dental Group, Matteo Magagnoli, AI Strategy and Program Director at A.P. Moller - Maersk A/S and Søren Christensen, Head of Artificial Intelligence at Bunker Holding Group.The conversation explores how AI can break down traditional barriers between technology teams and business teams, creating new opportunities for collaboration, innovation and problem-solving. The speakers discuss how AI tools can empower non-technical employees to work more effectively with data, develop solutions and contribute directly to business challenges. They also highlight the importance of providing employees with the right skills, guidance and access to AI while ensuring that its adoption delivers measurable business value rather than unnecessary costs or complexity.The discussion also examines how AI is reshaping organisational structures, responsibilities and the relationship between technical expertise and business knowledge. Key themes include data quality, governance, ownership, change management and the importance of maintaining human judgment and critical thinking when working with AI. The speakers emphasise that successful AI adoption is not simply about introducing new technology, but about bringing people, processes and technical capabilities closer together while retaining the expertise and human understanding that drive meaningful business outcomes.