Early Bird Savings: 10% off Tickets

EuroSTAR Conference

Europe's Largest Software Testing Conference.

2026

Agentic AI for Production-Grade, Domain-Specific Test Automation 

by

For the past few years, AI in testing has been dominated by impressive demos. A model generates a test script from a user story. A chatbot repairs a broken locator. A prompt magically produces automation code. 

But when enterprise teams attempt to deploy these capabilities into real CI/CD pipelines, they quickly discover a gap between possibility and practicality. 

Large Language Models (LLMs) are powerful, but they introduce serious challenges in production testing environments. They are inherently probabilistic; identical prompts can yield different results, undermining regression reliability. They can hallucinate syntactically correct but logically flawed automation steps. As test flows grow longer and more complex, models may lose context. Inference latency and cost can escalate. Model drift can alter behavior over time, requiring continuous revalidation. 

Small Language Models (SLMs), by contrast, are fast, stable, and cost-efficient. They perform exceptionally well in structured, domain-specific tasks. However, they lack the deep reasoning power required for complex, multi-step intent interpretation. 

So the real question is not “LLM or SLM?” 
It is: How do we combine both intelligently to build reliable AI-powered automation? 

The Hybrid Architecture: Precision Meets Reasoning

Production-grade AI testing requires architectural discipline. 

The most effective approach is a hybrid model strategy: 

  • Custom SLMs trained on domain-specific automation assets—keywords, UI definitions, reusable components, and workflow logic. 
  • LLMs are reserved for advanced reasoning tasks, such as decomposing high-level test intent into structured automation steps. 

This separation of responsibilities delivers determinism and efficiency for routine automation while preserving flexibility for complex reasoning. The result is faster inference, lower operational cost, improved stability, and reduced hallucination. 

Rather than relying on a single general-purpose model, a hybrid architecture assigns the right model to the right task. 

This architectural pattern can be applied to modern automation frameworks in general. At DWS, we have successfully implemented this hybrid model within TestArchitect (TA), demonstrating how domain-trained SLMs and reasoning-driven LLMs can operate together inside a structured automation ecosystem.  

TestArchitect supports a wide spectrum of platforms and technologies, including SAP, Salesforce, mobile native, mobile web, mobile hybrid, browsers, and desktop applications. Teams can create fully no-code automated tests in plain English, allowing business analysts, domain experts, and testers of all skill levels to contribute without coding barriers. This approach accelerates in-sprint automation, helping teams keep pace with rapid release cycles while ensuring high-quality testing across diverse applications. 

From Single-Model Scripts to Agentic AI 

Even hybrid models alone are not enough. True production-grade automation is not generated by a single model guessing a script; it is orchestrated. 

Agentic AI introduces a multi-agent system where specialized AI components collaborate across the testing lifecycle: 

  • Interpreting business-level test intent
  • Mapping intent to reusable automation actions 
  • Generating structured UI interaction steps 
  • Validating and refining codeless automation 
  • Executing across platforms 
  • Performing intelligent self-healing 
  • Classifying root causes of failures 
  • Predicting impact and suggesting remediation  

Instead of relying on a single probabilistic output, the system validates itself through structured collaboration. Each agent has a defined responsibility, reducing instability and improving reliability. 

Within CI/CD pipelines, intelligent test selection based on code impact analysis ensures that only relevant test suites are executed, reducing cycle time while preserving coverage. 

This marks a shift from AI as a script generator to AI as an orchestrated testing system. 

Within TestArchitect, this agentic hybrid model powers end-to-end intelligent automation, but the architectural principles are transferable to other enterprise automation environments. 

Data Engineering and Model Validation

The real differentiator in AI-powered testing is not model size. It is data quality and governance. 

For domain-intensive industries such as finance, healthcare, and energy, AI must align precisely with industrial workflows. This requires: 

  •  Structured automation datasets 
  • Clearly defined evaluation metrics 
  • Domain-specific fine-tuning 
  • Human-in-the-loop validation 

Domain experts play a critical role in guiding quality standards and ensuring AI-generated outputs meet enterprise expectations. 

To improve precision and consistency, Retrieval-Augmented Generation (RAG) is applied using structured project-specific automation data. Instead of relying solely on model weights, the system dynamically retrieves relevant context, existing action libraries, interface definitions, and project artifacts before generating tests. 

This approach reduces hallucination, preserves framework alignment, and improves maintainability. 

In TestArchitect implementations, this structured RAG strategy has enabled reliable AI-assisted test generation aligned with domain rules and enterprise governance standards. 

Deployment, Cost, and Sustainability

Enterprise adoption requires more than innovation; it requires sustainability. 

Custom SLMs can be hosted in private cloud or on-premises environments, ensuring that sensitive QA artifacts, requirements, test cases, and defect logs remain within organizational boundaries. Strategic collaboration with cloud-hosted LLMs minimizes unnecessary inference costs. 

Because SLMs are lightweight and CPU-friendly, organizations avoid GPU-heavy infrastructure and excessive cloud spending. Millisecond-level responses enable seamless integration into Agile and DevOps workflows. 

The hybrid agentic approach ensures that AI-driven automation remains: 

  • Secure
  • Cost-efficient
  • Predictable
  • Scalable

Without architectural discipline, AI in testing becomes unstable and expensive. With the right design, it becomes a long-term competitive advantage. 

Engineering the Future of AI-Driven Testing

AI will not replace automation frameworks. It will augment them. 

The organizations that succeed will not be those experimenting with prompts, but those engineering intelligent systems, combining hybrid models, agent orchestration, structured data pipelines, and rigorous validation. 

This hybrid, agentic AI architecture can be applied across automation ecosystems. Its successful implementation within TestArchitect demonstrates that production-grade, domain-specific, cost-controlled AI testing is not theoretical—it is achievable today. 

The future of testing belongs to teams that treat AI as an engineering discipline. 

Get Started with TestArchitect AI Today 

Author

Tuan Truong – Head of Test Architect Product Development

He leads the design and evolution of enterprise-scale test automation solutions, helping global organizations modernize quality engineering practices and deliver large-scale systems with confidence. With over 20 years of experience in software testing, automation architecture, and product engineering, Tuan specializes in integrating AI into practical testing workflows. His current focus is on designing production-grade AI systems that combine Custom SLMs, LLMs, and Agentic AI to create scalable, cost-efficient automation solutions. He works closely with enterprise teams to transform emerging AI capabilities into reliable, real-world testing systems. 

DWS are a participating in EuroSTAR Conference 2026 as a Gold Sponsor

How We Learned to Test Our RAG (and Accidentally Tested Our Content) 

by

In the last four years as a developer advocate at Qase, I’ve written more blog posts and LinkedIn articles than I can count. I’ve also reviewed hundreds of pieces from colleagues, checking whether the arguments hold, the examples work, and the claims are backed by evidence. 

I got better at it over time. My earliest articles are nowhere close to my latest ones, both in knowledge and in writing style. Naturally, I tried using LLMs along the way. I quickly found out that LLMs can’t be trusted to write on my behalf, and they certainly can’t be trusted to fact-check. They hallucinate confidently, they miss nuance, they produce text that sounds right but often isn’t. 

I did find one pattern that works: using LLMs as critics. I write, then I ask the LLM to poke holes in what I wrote. The diversity of their criticism is genuinely helpful. Even when they hallucinate in their feedback, it doesn’t matter, I still have to go through every point and decide what to keep. The hallucinations become noise I can filter, not errors that leak into my content. 

This was my comfortable setup for a while. Then my friend Anupam Krishnamurthy showed me something that changed how I think about content quality entirely. 

The Spark

Anupam and I co-own BeyondQuality, an open research community for software quality topics. He presented his research on evaluating RAG systems using a framework called Ragas. 

For those unfamiliar: RAG (Retrieval Augmented Generation) is a way to make LLMs answer questions from your own documents. You chunk your content, store it in a vector database, and when someone asks a question, the system retrieves the relevant chunks and feeds them to the LLM to generate an answer. Ragas is a framework that evaluates how well this works: does the answer stick to the retrieved content? Did the system retrieve the right content in the first place? 

Anupam built a RAG on the 37signals employee handbook and used Ragas to evaluate it. This was eye-opening. I started thinking about LLMs and content quality in a completely new way. 

From Curiosity to Production

My first experiment was pure fun: I built a chat interface to a Deming book I own. It worked, but there was no real business case. I know what Deming writes; I’d rather just re-read the book. 

The second idea had a real business case. At Qase, we have over a hundred blog posts, a Help Centre, and a customer support knowledge base. Every new piece of content operates on top of everything published before. If a new article contradicts something from six months ago, readers notice. It confuses people and erodes trust. 

This is a classical regression problem, just in content, not in code. A company with hundreds of published pieces has the same problem as a codebase with no tests: changes go out unchecked against existing behavior. Before this, the only check was someone reading a draft and trying to remember if it contradicts anything. With a hundred articles, that’s not realistic. With RAG, I could retrieve just the relevant pieces from the existing corpus and check the new draft against them automatically. Ragas then evaluates how well the retrieval and checking actually work. The feedback loop went from “hope someone remembers” to a 25-minute automated run. 

What Happened When We Tried It 

The first obstacle was building the evaluation itself. To test whether the system answers correctly, you need ground truths: the key claims and ideas from your content. I tried using several LLMs to extract these from our 106 blog posts. They all gave inconsistent results. The LLMs couldn’t agree on what the articles were actually saying. I ended up reading every article myself and writing down the key ideas manually. There is no shortcut here yet. 

Then I had to write evaluation questions. This turned out to be the same problem as writing tests after the code is already written: since you already know the system, you’re inclined to write tests that just confirm what’s there. My first questions were like that. They assumed knowledge of the content and led toward the answer. It took three iterations to learn to write questions the way a real person would ask them: short, simple, with genuine uncertainty about which way the answer goes. 

Once I had 240 questions and the evaluation was running, the results told a clear story. On broad, open-ended questions, the LLM stopped relying on the retrieved content and started answering from its own general knowledge. It sounded confident and correct, but it was no longer grounded in what we actually wrote. Ragas caught this. Without measurement, we would never have noticed. 

The whole evaluation for 240 questions costs $0.60 and takes 25 minutes — I’ve seen classical automated tests’ suites running for longer! 

And then the biggest surprise. For some questions, the system couldn’t find the right article to answer from. Not because the retrieval was broken, but because we simply hadn’t written about those topics well enough. We set out to test the RAG. We ended up finding holes in our own content. 

Where This Is Going

This research started with Anupam. Without his curiosity and his work at BeyondQuality, I would not have explored this direction at all. 

Today we have regression testing for our content: the blog, the Help Centre, the CS knowledge base. New drafts get checked against everything we’ve already published. What started as a contradiction checker is now also helping us find gaps in our existing content we didn’t know were there. 

If you want to see how RAG evaluation works hands-on, come to Anupam’s live demo at EuroStar in Oslo this June, 2026. He will walk through building and evaluating a RAG system step by step. And if the intersection of software quality and AI interests you, follow our work at BeyondQuality, where all research is published openly. 

I’ll be at the Qase booth throughout the conference. If any of this sparked your curiosity, come say hi. 

Author

Vitaly Sharovatov 

As a quality enthusiast, I believe that people should take pride in their work and companies should aim to produce high-quality products. I have spent the last 24 years in IT, focusing on engineering, quality assurance and mentorship. I am also a huge animal lover and have saved and raised more than 50 cats and dogs. 

QASE are a participating in EuroSTAR Conference 2026 as a Gold Sponsor

The 4 Pillars of Modern Testing: Building a Unified Ecosystem

by

In the pursuit of digital transformation, many QA teams are hindered by integration debt—the hidden cost of reconciling manual data and bridging visibility gaps between disconnected platforms. This debt is paid in the form of administrative overhead, reduced velocity, and the inherent friction that occurs when test management and test automation operate in separate silos.

To address these complexities, many organizations utilize the Inflectra ecosystem to bridge the gap between planning and execution. As a provider of software lifecycle management tools, Inflectra focuses on creating a single source of truth across the development pipeline. Within this framework, SpiraTest (test management) and Rapise (test automation) are engineered to function as a unified system, ensuring that automated execution remains tethered to original
business requirements.

1. Transitioning from Integration to Unification

Traditionally, QA has been bifurcated. Test management tools track requirements and manual progress, while automation engines function within independent environments. This disconnect creates three primary strategic risks:

  • The Transparency Gap: Stakeholders may see successful execution reports without understanding which specific business requirements have been validated.
  • Version Divergence: Automated scripts often evolve independently of the test plans they support, leading to deceptive results based on outdated logic.
  • Operational Inefficiency: Engineers frequently duplicate effort by recreating manual test steps in code because a shared source of truth does not exist.

2. Transitioning from Integration to Unification

There is a fundamental difference between two tools connected by an API and a truly unified ecosystem. In a unified environment, the automation engine is not an external add-on; it is a native extension of the management layer.

When automation is managed as a primary asset within the test management tool,
organizations achieve End-to-End Traceability. This allows a Project Manager to evaluate a high-level requirement and immediately view the precise execution logs and evidence that confirm its stability.

The Pillars of a Unified QA Workflow

To modernize a QA department, leadership should focus on four technical pillars that define a mature, unified strategy:

Pillar I: Requirement-Driven Automation
Instead of developing scripts in isolation, a unified system allows teams to derive automation directly from manual definitions. By using the manual test case as the structural blueprint, Rapise ensures that automation mirrors the original business intent. This alignment ensures that any change to a manual requirement in SpiraTest is immediately reflected in the automation gap analysis.

Pillar II: Centralized Orchestration
Automation provides the greatest ROI when it is accessible and autonomous. A unified system acts as a Command Center, enabling teams to schedule and remotely execute tests across global labs directly from the management interface. Orchestration ensures the automatic capture of screenshots, logs, and outcomes into a centralized, auditable trail.

Pillar III: Data-Driven Scalability
Professional maturity in testing involves moving beyond simple record-and-playback functions toward Parameterization and Parallel Execution. By defining complex data sets in SpiraTest and deploying them through Rapise scripts simultaneously, teams can expand test coverage exponentially without increasing their maintenance footprint.


Pillar IV: The Continuous Feedback Loop (CI/CD)
In a modern CI/CD pipeline, the test management tool serves as the definitive quality gate. When build completions trigger Rapise automations, the results feed directly into SpiraTest release dashboards. This creates a self-documenting loop that provides stakeholders with a data-backed assessment of risk before code reaches production.

Leadership Perspective: Quality as a Discipline

True software excellence happens when strategy and execution are unified. By adopting a streamlined workflow, organizations reduce toolchain complexity and eliminate the
labor-intensive effort required to synchronize disparate platforms. The result is a more resilient QA process where quality is a continuous, intelligent discipline rather than a final checkpoint.

Summary for QA Executives

If your team spends more time managing tools than testing software, your architecture is suffering from integration debt. A unified system like SpiraTest and Rapise goes beyond task automation; it synchronizes your entire quality strategy, ensuring that every automated action serves a documented business goal.

Author

Adam Sandman

Adam Sandman is a visionary entrepreneur and a respected thought leader in the enterprise software industry. As the Founder and CEO of Inflectra Corporation, Adam has dedicated his career to revolutionizing how businesses approach software development, testing, and lifecycle
management.

Under Adam’s leadership, Inflectra has become a global provider of award-winning solutions, from SpiraTest’s powerful test management and flexible automation of Rapise to SpiraTeam’s end-to-end traceability. He has led Inflectra’s suite of software to grow into a global standard that empowers teams across the world to deliver high-quality software efficiently and collaboratively. His deep technical expertise, combined with a passion for innovation, has
positioned him as a trusted voice in the field, influencing trends and shaping best practices for agile development and quality assurance.

Adam is known for his engaging presentations at industry conferences, where he shares insights on topics such as automation, project management, and emerging technologies. His ability to translate complex concepts into actionable strategies has earned him a reputation as an effective educator and mentor.

Beyond his technical acumen, Adam is committed to fostering a culture of inclusivity and collaboration within the tech community. Through thought-provoking blogs, webinars, and public speaking engagements, he inspires professionals worldwide to adopt forward-thinking approaches to software development and testing.

When he’s not at the helm of Inflectra, Adam enjoys exploring the latest advancements in technology, mentoring up-and-coming tech leaders,

Inflectra is participating in EuroSTAR Conference 2026 as a Gold Sponsor

AI is accelerating testing, but judgment still sits with people. 

by

GenAI in Testing: Acceleration Without Abdication 

In many software teams today, generative AI is becoming part of everyday engineering work. 

Within testing, it is being used to turn requirements into draft test outlines, generate realistic data variations, structure API and automation tests, refactor brittle scripts, and make sense of complex logs. What began as curiosity is now influencing how testing work is approached. 

Industry research reflects this shift. The Stack Overflow Developer Survey 2025 reports widespread use of AI tools across development workflows (https://survey.stackoverflow.co/2025/).

At the same time, the Capgemini World Quality Report highlights AI-driven testing and intelligent automation as priorities for quality leaders (https://www.capgemini.com/insights/research-library/world-quality-report/). 

AI is shaping both daily testing practice and broader quality strategy.  The important question is not whether we use these tools. It is whether we understand their limits. 

AI as Support, Not Authority 

Generative AI can suggest alternative test conditions, propose data sets, draft automation code, explain unfamiliar system behaviour, and assist with log analysis. Used carefully, it reduces friction and expands exploration. 

But it does not understand context. 

It does not evaluate risk. 

It does not decide what matters most. 

Testers do. 

AI can produce output. Testers determine whether that output addresses genuine system risk. Are edge cases meaningful? Is automation maintainable? Does the proposed coverage reflect realistic user behaviour? Has bias crept into assumptions? 

The fundamentals of testing remain unchanged: risk-based thinking, analytical depth, system modelling and critical reasoning. Working alongside AI does not replace these skills. It makes the gaps more visible when they are missing. 

Testing AI Systems Is a Different Discipline 

The shift is not only about using AI in testing. Many teams are also validating AI-enabled systems.  

These systems behave differently from traditional deterministic software. Outputs may be probabilistic. Behaviour can evolve as training data changes. Bias may appear in subtle ways. Explaining outcomes can be more complex than reproducing them. 

Testing in this environment requires additional competence. Understanding data quality, model behaviour, bias detection and the implications of non-deterministic results is increasingly part of the tester’s remit. 

The certification schemes developed by the ISTQB reflect this evolution. Specialist pathways now address AI and machine learning testing alongside established foundations in structured test design and risk analysis. 

Independent examination for these certifications is delivered globally by organizations such as iSQI GmbH. Further information on certification pathways can be found at https://www.isqi.org. 

Certification does not replace experience. It does not manufacture judgement. What it offers is structured knowledge, shared language and a clearer route for professional development in an expanding discipline. 

Learn more about Testing with Generative AI and Testing AI systems 

A Balanced View of Certification 

Certification has long been debated in the testing community. That debate is appropriate. 

Testing is contextual. It involves thinking, adapting and responding to nuance. No syllabus can capture every scenario encountered in practice. 

However, structure has value. Shared terminology reduces misunderstanding. Defined knowledge areas help organisations benchmark capability. Independent assessment provides transparency in a global profession where skills must be portable. 

The aim is not to standardise judgement. It is to support professional growth. 

Speed Is Not Insight 

AI unquestionably increases speed. Draft artefacts can be produced quickly. Data sets can be generated at scale. Logs can be summarised in seconds. 

The danger lies in mistaking output for understanding. 

Generating hundreds of test cases is easy. Determining whether they address real risk is harder. Automation can execute rapidly. Interpreting results in context still requires experience. 

When strong testing fundamentals are combined with informed AI use, productivity improves without compromising quality. When fundamentals are weak, AI amplifies the weaknesses. 

The testing profession has navigated similar shifts before. From manual execution to automation. From sequential delivery to agile development. From monolithic systems to distributed architectures. Each transition required adaptation while preserving core principles. 

Generative AI is another transition. 

The tools will continue to evolve. The need for informed human judgement will not. 

Quality has always depended on professionals who understand risk, context and consequence. That remains true. 

The real conversation is not whether AI belongs in testing. It is whether we are developing testers who can use it with discipline and intent.  

Author

Debbie Archer 

Debbie Archer is Managing Director of iSQI Limited and Vice President for Business Development at iSQI Group. With more than 20 years of experience in learning and development, she has held senior positions, including Director of Global Channel Partners at the BCS, The Chartered Institute for IT. She is actively involved with ISTQB® (International Software Testing Qualifications Board), supporting the development of globally recognised software testing qualifications. 

iSQI are a participating in EuroSTAR Conference 2026 as an Exhibitor. Join us at EuroSTAR Conference EXPO in Oslo 15-18 June 2026.

From Oxymoron to Reality: How to Execute Manual Tests Automatically with AI 

by

How does the automatic execution of manual tests speed up test cycles? 

On the surface, “automatic execution of manual tests” sounds like an oxymoron. The industry has always operated on a binary: manual means a human tester; automatic means a script, a framework, and an automation engineer. For decades, “manual” has been synonymous with human intervention. 

But agentic AI is turning this contradiction into a practical capability. At the heart of this shift is Lynqa, an AI agent designed to run manual tests directly on the GUI. It understands test procedures, navigates the application interface, verifies results, and delivers a detailed execution report without a single line of automation code. 

The Sprint Bottleneck: Why “Manual” Needs a Boost 

Every QA team knows the pattern: as the sprint nears completion, the volume of manual checks piles up, testers are stretched thin, and feedback to developers slows to a crawl. This is the “Manual Testing Debt.” 

The usual responses, hiring more testers or rushing to automate new features, rarely work within a single sprint. Traditional test automation simply takes too long to develop alongside the feature it’s meant to validate. 

This is where GUI AI agents change the game.  

What Are GUI AI Agents? 

GUI AI agents interact with software by visually interpreting screen interfaces and operating a virtual mouse and keyboard. Instead of relying on backend APIs, these AI agents execute complex digital workflows exactly as a human user would: 

  1. GUI-Based Actions: It interacts with the graphical interface exactly as a human would, opening browsers, scrolling through menus, filling in fields, and navigating complex clients such as ERP systems. 
  1. Visual Perception: It monitors progress by analyzing the screen at each step, identifying the areas affected by an action or verification. 

The Feedback Loop: Thought and Action 

The agent operates in a continuous feedback loop that combines reasoning with action: 

  • Interpretation: It reads the test steps and builds a logical action plan. 
  • Perception: It analyzes the current screen to decide which interaction (click, type, drag) to perform. 
  • Monitoring: After every action, it checks the UI state to verify the expected result. 
  • Communication: If the agent encounters an ambiguity or a high-risk step, it pauses to ask the user for clarification before proceeding. 

This means the agent doesn’t blindly follow a script; it reacts to the application’s actual behaviour. 

Lynqa: Agentic AI Inside Your Test Management Tool

While general-purpose AI agents are appearing in consumer chatbots, Lynqa for Jira is purpose-built for the testing ecosystem. By integrating directly into Jira/Xray, it transforms how teams handle manual workloads within a sprint. 

How it speeds up the cycle: 

  • Zero Scripting Overhead: If you have a written test case or a clearly defined User Story, the agent can execute it immediately, without waiting for an automation engineer. This enables “Day 1” testing of new features. 
  • Adaptive Execution: When a button moves, a CSS class is renamed, or a color shifts, traditional automation breaks. The AI agent sees the “Submit” button regardless of underlying code changes, removing the maintenance tax that halts progress. 
  • Analysis-Ready Reporting: The agent doesn’t just return “Pass” or “Fail.” It provides step-by-step screenshots and a detailed failure analysis, giving testers actionable information immediately. 

Start your experience with Lynqa today! We offer 10 free credits upon installation, or you can book a demo with our team. Find more information here.  

A Collaborative Future: The QA “Tribe” 

At EuroSTAR, we talk about “finding your tribe” and sharing the passion for quality. AI agents aren’t here to replace human testers; they’re here to empower the tribe. 

The most effective teams use agents like Lynqa as extra hands. While the agent handles repetitive GUI-level checks and initial failure analysis, human testers reclaim time for the work that matters most: exploratory testing, risk analysis, and strategic quality coaching for the development team. 

The tester’s role evolves from executor of steps to orchestrator of agents. As we look toward the 2026 EuroSTAR Conference, the question is no longer whether manual tests can be automated; it’s how quickly your team can leverage agentic AI to keep pace with modern delivery. 

Author

Bruno Legeard,  Smartesting

Bruno Legeard heads up the AI Lab at Smartesting (publisher of Lynqa). An expert in AI for software testing, he is one of the authors of the ISTQB CT-GenAI certification “Testing with Generative AI.” 

Meet the Lynqa team at the Smartesting booth during the EuroSTAR 2026 Conference to see agentic testing in action! 

Smartesting are Exhibitors at EuroSTAR 2026. Join us at EuroSTAR Conference in Oslo 15-18 June 2026

Testing at Its Best: Inside the EuroSTAR 2026 Programme 

by

When we started shaping the EuroSTAR 2026 programme, one question kept coming back to us: 

What does testing at its best really look like today? 

To help answer that question, Programme Chair Elmar Jürgens (CQSE GmbH, Germany) worked alongside our dedicated Programme Committee, bringing together their expertise, insight, and passion to create a programme that is both practical and inspiring. 

The committee members – Fiona Østensvig (Gritera Quality, Norway), Willem Keesman (Sopra Steria, Netherlands), Richard Seidl (Consulting GmbH, Germany), and Sophie Küster (cronn GmbH, Germany) -each bring their own unique experience.  

Together, they’ve designed a programme that tackles the real challenges testers face today, offers opportunities to learn and grow, and celebrates the collaboration that makes our community so strong. 

AI in Testing: Practical, Thoughtful, and Human-Led 

AI is everywhere in the EuroSTAR 2026 programme, but what’s refreshing is how it’s being discussed. Rather than “AI will replace testers” narratives, the sessions focus on practical collaboration between humans and machines.  

Sessions explore how testers are already using AI in their work and what’s required to do this well. Tutorials such as AI-Driven API Test Automation and talks on training and guiding Large Language Models focus on real applications, not promises of full automation. 

Other sessions take a more reflective approach, asking important questions about trust, bias, and responsibility when AI becomes part of the testing process. Across these talks, one message is consistent: AI can enhance testing, but it still relies on human judgement, context, and critical thinking. 

Testing at its best means understanding both the power and the limits of these tools. 

Context, Risk, and Understanding the System 

Another strong theme running through the programme is context. 

Sessions on context engineering, exploratory testing, and risk-based approaches reinforce the idea that quality cannot be separated from the environment in which software exists. These talks focus on helping testers understand why something matters – not just how to test it. 

The Human Side of Quality 

Alongside technical depth, the EuroSTAR 2026 programme continues to highlight the human skills that support great testing. 

Talks exploring psychology in testing, communication and collaboration, and mental health in remote testing teams. Quality depends on conversations, relationships, and the confidence to ask difficult questions. 

As testing tools become more advanced, these skills become even more valuable. 

EuroSTAR keynotes are designed to challenge thinking and spark meaningful discussion — and the 2026 keynote line-up does exactly that, bringing insight, experience, and inspiration to the main stage. 

The Jobs Not Taken 

Dona Sarkar — Microsoft, USA 
Dona Sarkar encourages us to embrace experimentation in our careers and work. Drawing on her role as Chief Troublemaker for Microsoft’s AI and Copilot Extensibility Program, she discusses how curiosity and bold experimentation can unlock growth, creativity, and impact. Explore the session.

Do It For the Plot 

Sanne Visser — Bartosz, The Netherlands 
Sanne Visser explores resilience, curiosity, and consistent effort in testing. Using software projects as a metaphorical story, she shows how engaging fully with challenges — even uncertain ones — fosters growth, learning, and meaningful progress in our daily work.  Explore the session.

Three Decades of Software Testing – From Startup to Global Leader 

Wolfgang Platz — Katharo Ventures, Austria 
Wolfgang Platz reflects on his journey from early software development to founding Tricentis, Austria’s first unicorn. He shares lessons from three decades in testing, including innovation, leadership, and how testing practices have evolved to meet the challenges of modern technology.  Explore the session.

With Great Trust Comes Great Responsibility 

Michael Kutz — REWE digital, Germany 
Michael Kutz examines the role of trust in high-performing teams and software quality. Drawing on his experience establishing agile practices at REWE digital, he highlights how autonomy, accountability, and responsibility underpin successful testing and development cultures.  Explore the session.

What “Testing at Its Best” Means to Us 

When we look at the EuroSTAR 2026 programme as a whole, a few things stand out to us.

AI is now part of everyday testing — and learning how to use it thoughtfully and responsibly is something we all need to consider.

Context, risk, and truly understanding the system still sit at the heart of quality.

And human skills? They’re not secondary. They’re essential.

For us, testing at its best isn’t about moving faster for the sake of it, or automating everything we can. It’s about making informed decisions, understanding what matters, and applying our skills where they have the greatest impact.

Looking Ahead to EuroSTAR 2026 

EuroSTAR has always been about bringing the testing community together — to learn from each other, challenge ideas, and share experiences. The 2026 programme continues to do so, offering a mix of deep technical learning, human-centred topics, and opportunities for discussion. 

We’re looking forward to welcoming the community to the Nova Spektrum in Lillestrøm and exploring what testing at its best can look like – together. 

Autonomous Testing with AI: Separating Hype from Real-World Value

by

Autonomous testing is one of those phrases everyone nods at, but few people can clearly explain. Slide decks talk about systems that write their own tests, fix failures automatically, and keep pipelines green without anyone watching. Somewhere along the way, teams start wondering if they are falling behind simply because they don’t have “autonomy” yet.

Most of that anxiety comes from hype, not reality. The real question isn’t whether testing can become autonomous. It’s where autonomy actually helps teams do better work, and where it quietly introduces new risks.

Autonomy is a spectrum, not a switch

Testing doesn’t suddenly flip from human-driven to autonomous. It evolves in layers.

Most teams start with automation that focuses on execution. Scripts run faster than humans, but people still decide what gets tested, how coverage is shaped, and whether results can be trusted. This is familiar territory.

AI assistance adds another layer. Test creation speeds up, failures are grouped instead of dumped into long reports and maintenance effort drops. These gains are real, but they are still reactive. Humans remain firmly in control.

Autonomy begins only when a system can take on limited decision-making, within boundaries defined by the team. Not creative judgment, but practical judgment. Deciding which tests are relevant for a specific change. Flagging failures that look like noise rather than regressions. Noticing flows that are becoming so fragile they can no longer be trusted. This is usually where expectations and reality part ways. What is sold as autonomy often turns out to be faster automation with a new label.

Why Hype Breaks Down in Real Environments?

Autonomous testing struggles when context is missing.

AI does not understand business importance on its own. It cannot tell which workflow carries regulatory exposure or which release is under executive scrutiny. Without that context, decisions become statistical guesses rather than informed choices.

Self-healing is another frequent weak spot. Updating locators can keep tests running, but it can also hide changes in behavior that actually matter. Pipelines look stable while confidence quietly erodes.

A green pipeline is comforting. A trustworthy pipeline is far more valuable. Autonomy that masks risk does more harm than good.

Where AI Genuinely Helps?

Used well, AI shines at recognizing patterns humans don’t have time to see.

It can connect failure signatures across pipelines and environments. It can reveal tests that only break under specific data conditions. It can point out entire suites that consume time without improving coverage. It can highlight areas where manual testing consistently finds issues automation misses.

This doesn’t replace testers. It sharpens their judgment.

One of the most underrated benefits of AI-assisted testing is how it changes team discussions. Less time is spent arguing about flaky failures. More time is spent talking about system behavior and risk. For many teams, that shift alone justifies the investment.

Autonomy Needs Boundaries to be Trusted

The most effective autonomous testing systems are deliberately limited.

They don’t decide what quality means; teams do.

They don’t invent test strategies. They optimize what already exists.

They don’t operate silently. They explain their reasoning.

Explainability is non-negotiable. If a system skips tests or classifies failures without showing why, teams will override it every time. Accuracy matters, but transparency is what builds trust.

Autonomy is adopted when people understand not just the outcome, but the reasoning behind it.

From Scripts to Systems

Traditional automation treats tests as isolated scripts. Autonomous testing treats them as parts of a larger system.

That system understands dependencies between services, data, environments, and user flows. It recognizes that a login failure ripples across dozens of downstream tests. It understands that a configuration change in one region shouldn’t invalidate results everywhere else.

This shift is subtle, but important. Autonomy works not because AI is smarter than humans, but because it can track complexity humans can’t reasonably keep in their heads.

This way of thinking is increasingly reflected in how enterprise platforms are being designed. At ACCELQ, autonomy is treated less as an end state and more as a support system for decision-making at scale. The emphasis is on observing system behavior, correlating signals across pipelines, and making change easier to understand rather than hiding it behind execution.

Capabilities such as adaptive test generation, intelligent handling of change, and agent-style execution through ACCELQ Autopilot are used to reduce noise and maintenance while keeping teams firmly in control of intent and strategy. Autonomy, in this model, is not about removing oversight. It is about making complex testing environments more transparent as systems evolve.

How Teams Should Evaluate Autonomy Today?

A simple way to cut through autonomous testing claims is to ask a few uncomfortable questions. What decisions does the system make without human input? What signals drive those decisions? How does it behave when signals conflict? And how easy is it to override decisions and learn from them?

Vague answers usually point to cosmetic autonomy.

Real autonomy shows up quietly. Fewer late-night reruns. Fewer ignored failures. Fewer production surprises. It reduces friction without asking for attention.

The Future is Assistive, Not Absent

The strongest testing organizations aren’t trying to remove human judgment. They are trying to protect it.

AI is well-suited to repetition, correlation, and scale. Humans are better at intent, ethics, and trade-offs. Autonomous testing works when it creates space for thinking, not when it pretends thinking is no longer needed.

The real value isn’t in replacing testers. It’s in freeing them from work that hides risk instead of revealing it.

That’s the point where autonomy stops being a promise and starts being useful.

Author

Geosley Andrades, Senior Director,  ACCELQ

Geosley is a Senior Director, Product Evangelist and Community Builder at ACCELQ, leading global AI-driven, no-code test automation initiatives alongside product strategy and go-to-market programs. With nearly 18 years of cross-industry experience, he helps enterprises rethink how software quality is built, validated, and scaled for real-world impact. A strong advocate for intelligent, autonomous testing at enterprise scale, Geosley actively shapes ACCELQ’s vision through competitive analysis, analyst engagement, and forward-looking research—driving simpler, more reliable, and sustainable automation for modern digital ecosystems.

ACCELQ are Exhibitors at EuroSTAR 2026. Join us at EuroSTAR Conference in Oslo 15-18 June 2026

How Agentic AI is Transforming Software Delivery Lifecycles 

by

Software delivery has always been a balancing act between speed, quality, and risk. As enterprises adopt cloud-native architectures, DevOps, and continuous delivery models, that balance is becoming harder to maintain. Traditional automation and AI tools help — but they still rely heavily on human direction. 

Agentic AI introduces goal-driven intelligence into the software delivery lifecycle (SDLC), enabling systems to adapt, recommend, and act within defined enterprise guardrails. Rather than replacing human decision-making, agentic AI augments teams with continuous intelligence and policy-aware automation. The result is faster releases, higher quality software, and delivery pipelines that continuously optimize themselves. 

In this blog, we explore what agentic AI is, how it applies to software delivery, and why it represents the next evolution of enterprise DevOps. 

What is Agentic AI in Software Delivery? 

Agentic AI refers to artificial intelligence systems designed to act autonomously in pursuit of defined goals, rather than simply responding to prompts or executing predefined rules. 

In the context of software delivery, agentic AI systems can 

  •  Analyze delivery data across tools and teams. 
  • Make context-aware recommendations and trigger actions with minimal manual intervention, while remaining aligned to enterprise policies and approval models. 
  • Take coordinated action across the SDLC (planning, testing, release, operations) by leveraging a unified platform that connects delivery data, workflows, and governance in one system of record.
  • Learn and adapt based on outcomes. 

Unlike traditional AI, which primarily assists humans with insights or recommendations, agentic AI acts as an intelligent participant in the delivery process. 

The Role of Agentic AI Across The Software Delivery Lifecycle 

Agentic AI doesn’t replace DevOps teams — it augments them by operating continuously across every phase of delivery. 

Planning and Prioritization 

Agentic AI can analyze backlogs, historical delivery data, and business objectives to recommend optimal sprint scope, identify high-risk dependencies, and dynamically reprioritize work based on changing conditions. Instead of static planning cycles, teams gain adaptive planning intelligence that evolves in real time. 

Development and Build Automation 

During development, agentic AI agents can 

  • Detect code patterns linked to future defects. 
  • Optimize build pipelines based on performance trends. 
  • Flag architectural risks earlier in the process. 

This proactive intelligence reduces downstream rework and accelerates time to value. 

Continuous Testing and Quality Management 

Testing is where agentic AI delivers some of its greatest impact. AI agents can can intelligently prioritize and optimize test execution based on code changes, historical risk patterns, and release context—reducing redundancy while increasing confidence in release readiness. This leads to faster feedback loops and higher confidence in release readiness. 

Deployment and Release Optimization 

Agentic AI enables smarter, safer deployments by selecting optimal deployment windows, monitoring live performance and user impact, and detecting performance or reliability thresholds in real time and initiating guided remediation workflows. The result is a more resilient release process with fewer disruptions. 

Post-Release Learning and Optimization 

Unlike traditional automation, agentic AI continues learning after release by analyzing: 

  •  Customer feedback 
  • Production incidents
  • Delivery performance metrics

Those insights feed back into planning and execution, creating a continuously optimizing delivery lifecycle—where insights from production, quality, and delivery performance feed back into planning and execution with full transparency. 

Why Agentic AI Matters For Enterprise DevOps Teams 

Enterprise DevOps environments are complex — multiple tools, distributed teams, hybrid architectures, and competing priorities. Research from Google’s DORA team has consistently shown that high-performing DevOps organizations deploy software more frequently, recover faster from incidents, and maintain higher reliability than their peers. As delivery complexity increases, agentic AI helps teams manage that complexity at scale by enabling autonomous decision-making across the software delivery lifecycle. 

Google’s DORA research shows that high performance in deployment frequency, lead time, failure rate, and restoration time strongly correlates with effective DevOps strategies. 

Key benefits include: 

  •  Faster delivery cycles through autonomous decision-making 
  • Improved quality with predictive defect detection 
  • Reduced operational risk via real-time monitoring and response 
  • Lower cognitive load on teams, freeing humans to focus on innovation 

Rather than replacing human expertise, agentic AI allows teams to operate at a higher strategic level. 

How Agentic AI Fits into The Modern Software Delivery Stack 

Agentic AI works best when embedded within an integrated, enterprise-grade software delivery platform that connects planning, development, testing, and operations data. 

A unified platform enables AI agents to see the full delivery value stream, correlate signals across tools, and take informed, context-aware actions. This is where modern, AI-enabled delivery platforms play a critical role. 

Real-World Use Cases of Agentic AI in Software Delivery 

Organizations adopting agentic AI are already seeing tangible results, including: 

  • Automatically identifying release risks before production 
  • Reducing test execution time while improving coverage
  • Accelerating recovery from incidents without manual intervention
  • Optimizing delivery flow across large, distributed teams  

These outcomes are especially valuable for enterprises managing mission-critical applications and frequent releases. 

The Future of Software Delivery is Autonomous and Intelligent 

As software delivery continues to accelerate, static automation and manual oversight will no longer be enough. Agentic AI represents the next step forward — enabling delivery systems that think, act, and improve continuously. 

For organizations looking to improve speed, quality, and resilience at scale, agentic AI isn’t a future concept — it’s a competitive necessity. 

By combining agentic AI with an integrated software delivery platform, enterprises can unlock a new level of delivery performance and innovation. 

Deliver Software Faster—With Intelligence Built in 

OpenText brings agentic AI to software delivery by connecting planning, development, testing, and operations into a single intelligent platform. See how autonomous insights and actions can help your teams deliver higher-quality software at speed. 

For more information about the OpenText Core Software Delivery Platform, Visit us at EuroSTAR 2026 on Stand 5 or click here. 

Author

Gabriel Martinez, Director of DevOps Product Marketing, OpenText 

Gabe’s enthusiasm for tech is only matched by his passion for marketing it. With a 15-year sprint through DevOps, Cloud Management, and Application Security, he’s played a key role at Broadcom/VMware, CloudBees, Electric Cloud, and CA Technologies. Currently, Gabe is the Director of Product Marketing for ADM and DevOps at OpenText, leveraging storytelling and strategy with creativity and data-driven precision to orchestrate growth and innovation. 

OpenText are Exhibitors in this years’ EuroSTAR Conference EXPO. Join us in Oslo 15-18 June 2026.

part of the