Quality Engineering with GenAI and Agentic AI

Quality engineering is about team members and their stakeholders taking joint responsibility to deliver IT systems with the right quality at the right moment. This involves all lifecycle activities. Some people think quality engineering is just a fancy name for testing, but it is much more. In this module you will read about a wide range of quality engineering activities and how GenAI can perform these activities as one of the crossfunctional team members.

People in a cross-functional team assume multiple roles to perform a wide range of tasks. This way, the team is flexible. Team members share the responsibility to complete all tasks on time.
On this website, we use the term quality engineer to refer to individuals who are part of a cross-functional team and capable of performing various tasks. The term does not denote a specific job title but serves as an umbrella term for those involved in delivering the right level of quality at the right time.
Quality engineering needs to be amplified by applying GenAI, other tools and smart approaches to ensure that every task within the IT delivery lifecycle is carried out in a balanced manner, concerning speed, quality, and business value. In today’s fast-moving world, the quality engineering conundrum of what to do and where to focus is magnified multifold. The convergence of physical with cyber has added another layer of complexity to quality engineering. The product strategy is shifting from building clearly defined products to building connectable eco-systems. Connections between such eco-systems have opened new opportunities as well as new vulnerabilities.

Amplified quality engineering is used to achieve an accelerated and optimized level of quality of the IT delivery process by using an intelligent approach to all activities in the software development life cycle.

Quality Engineering with GenAI Is Context-Dependent

Quality Engineering with GenAI does not manifest uniformly across organizations. Its shape, impact, and required capabilities depend strongly on the organization’s level of AI adoption. The same GenAI technology can serve as a productivity tool in one organization, a decision-shaping mechanism in another, or be part of an autonomous execution landscape in a more advanced environment. Therefore, understanding the maturity context is essential before defining the role of Quality Engineering.

We introduced three fundamental ways of using GenAI:

As a tool based on prompting – supporting individual productivity (e.g., generating tests, code, or documentation).
As an assistant with a GenAI-based tool – structurally embedded in workflows, influencing analysis and recommendations.
As an autonomous or agentic actor with Agentic AI – executing actions within defined boundaries.

These three usage patterns form the foundation for the five organizational levels of AI adoption and usage:

Level 1 – Tool-Based Usage: GenAI is used ad hoc by individuals. Quality Engineering (QE) remains largely traditional in processes and activities.
Level 2 – AI-Assisted Delivery: GenAI is integrated into pipelines and workflows. QE shifts toward validating automation and ensuring automated control.
Level 3 – AI-Augmented Decision Systems: GenAI influences operational decisions. QE becomes responsible for explicit risk governance and transparency.
Level 4 – Agentic Systems: GenAI executes delegated actions within guardrails. QE designs autonomy boundaries, escalation mechanisms, and observability.
Level 5 – Quality-by-Design in AI Ecosystems: GenAI is embedded strategically across the enterprise. QE evolves into a trust architecture discipline, shaping governance and risk appetite at the organizational level.

These levels do not represent a mandatory growth path. Rather, they describe fundamentally different operating models. Each level implies a different risk profile, a different distribution of responsibility between humans and machines, and a different approach to quality engineering.

The Need for Speed

In a world where systems are deeply interconnected and subject to continuous change, the challenge remains to uphold quality standards while delivering business value at the right moment. An Agile mindset and DevOps culture have brought us far in terms of speed and adaptability. However, the stakeholder demand for acceleration persists. The next step lies in expanding automation—not only in operational tasks like test execution, but also in knowledge-intensive activities. Emerging practices such as pair programming with GenAI coding assistants illustrate how intelligent automation can support teams beyond traditional boundaries. These new possibilities enable IT delivery teams to reach the next level of speed in the evolution of the cycle time of quality engineering and testing.

Figure. Evolution of cycle time of quality engineering and testing.

This figure illustrates the increasing speed at which teams perform tasks. It’s important to remember that speed is not the only measure of success. Maintaining a balance between quality and time is essential, as described in the definition of quality engineering that says “the right quality at the right moment”.

Providing the Right Information About Quality and Risks

The goal of quality engineering is to continuously deliver IT systems with the right level of quality at the right time. As the pace of IT delivery accelerates, the window for gathering information
about the quality and related risks has narrowed. In the past, human testers often relied on personal expertise and gut feeling to assess quality. Today, we aim to augment that process with
modern GenAI-powered tools. However, these tools lack human intuition, so we must rely on other means. This is where traditional practices, such as test design techniques, become increasingly
relevant to ensure that quality levels remain measurable, traceable, and actionable in this new era.

This module gives insight into how GenAI can support the quality engineering activities to maintain the right level of quality and still increase the delivery speed.
This module contains many new insights, building on top of the work that was done by the authors of the TMAP book Testing in the digital age: AI makes the difference [Ven 2018].

The following building blocks are relevant within this module, Quality Engineering with GenAI and Agentic AI: