Gespräch mit dem Top-Architekten Christian Ciceri: Überwindung des traditionellen Denkens - In den Augen eines Spitzenarchitekten ist es nicht die technische Fähigkeit, die die Obergrenze einer Karriere bestimmt.
Against the backdrop of the rapid development of software development and the collision with the wave of artificial intelligence, the role and methodology of software architects are experiencing an unprecedented test and change.
In the past, the core tasks of architects mainly focused on system design, module division, and technological decisions. Emphasis was placed on stability, maintainability, and mastery of the technology stack. However, with the spread of cloud - native architectures, microservices, large distributed systems, and low - code/no - code platforms, the complexity of software systems has increased exponentially. Architects are no longer only faced with technical selection problems but also with the challenge of maintaining the health of the architecture and the efficiency of the team in an environment of rapid iterations and continuous deployment.
At the same time, the rise of artificial intelligence has given software development previously unknown tools and capabilities. Technologies such as automatic code generation, intelligent testing, and AI - supported design enable some traditional architecture tasks to be quickly completed by algorithms. Architects no longer have to draw all dependency diagrams themselves or manually analyze performance weaknesses but can quickly discover potential problems and optimization opportunities with the help of AI. Although this change increases efficiency, it also brings new questions: Decision - making power, system understanding, and technical judgment still strongly depend on human experience and insight. How can one ensure that AI tools support but not replace? This is the central question that architects must face.
It can be said that software architects are in a transformation phase in which technical skills, business knowledge, and data - driven decisions are interconnected. In an environment of rapid iterations, they must master new technologies, understand the development patterns of complex systems, and at the same time lead the team to develop a common vision and an architecture culture.
Recently, InfoQ conducted an in - depth interview with Christian Ciceri, co - founder of Apiumhub and a well - known software architect. Readers will be guided by the practical experiences of front - line architects to deeply discuss the idea of "measurable, evolvable architectures" and the impacts of AI and modern software development tools on architecture practice. This interview not only provides an overview of Ciceri's career but also offers rich knowledge about architecture practice so that readers can understand how to maintain the architecture quality and the adaptability of the team in a rapidly changing technological environment.
Ciceri's career path is representative: He has accumulated his experiences in front - line software development and architecture and has experienced common challenges in large enterprises such as lack of flexibility, long delivery cycles, and low process efficiency. In 2014, he founded Apiumhub in Barcelona together with Evgeny Predein, aiming to closely combine agile methods and software architecture with the core of business operations. In his long - term practice, Ciceri gradually developed the idea of "measurable, evolvable architectures", which he summarized in his book "Software Architecture Metrics". He emphasizes that building a solid and adaptable system not only improves the quality of software delivery but also ensures that the system grows in sync with business requirements.
In the interview, Ciceri elaborated in detail on the concepts of "observability" and "architecture governance". He pointed out that the quality characteristics of a system at runtime only represent a part of the overall quality. Real architecture governance requires continuous monitoring of all software properties. With the help of fitness functions in evolvable software architectures, teams can monitor the health of the architecture in real - time and recognize early signs of architecture degradation such as decreasing development speed, increasing errors, or performance problems.
Another highlight of the interview was Ciceri's insight into the role of AI in software architecture. He made it clear that AI can help in analyzing indicators and suggesting improvements but cannot replace human judgments and decisions. He emphasizes: "Only when driven by humans can AI become a real productivity aid in software design instead of a replacement." Regarding the currently AI - generated architecture suggestions, he still sees them as "helpers" rather than "partners" and warns architects to remain rational when accepting AI.
From Architect to Founder
InfoQ: Good day, Mr. Ciceri. From your book, I know that you are an advocate of "measurable, evolvable architectures". Can you first tell us how you went from being a front - line software architect to the co - founder of Apiumhub and how the writing of "Software Architecture Metrics" came about?
Christian: I owe my professional development to many years of practical experience in software development and architecture in large enterprises. During this time, I have experienced typical industry challenges such as lack of enterprise flexibility, long delivery cycles, and low process efficiency.
In 2014, I founded the company Apiumhub in Barcelona together with Evgeny Predein. Our core goal was to put agile methods and software architecture at the center of business operations.
Over time, I realized that to create long - term value, one must focus on measurable and evolvable architecture designs. Only a solid and adaptable system enables teams to deliver high - quality software and ensures that the software grows with business requirements. This idea finally led me to write the relevant chapters in "Software Architecture Metrics".
InfoQ: In your book, I noticed that many architects focus on "vision", while you focus more on "measurement" and "indicators". Why is measurability so important in modern software architecture?
Christian: Indicators and measurements are effective means to make debates objective. However, it is important to note that the team's common architecture vision is still a very important part of the daily work of software architects.
InfoQ: AI tools are already automating parts of design, coding, and testing, but architecture still depends on human judgments. Which aspects of architecture design can actually be improved by AI? Which should still be led by humans? How has your architecture philosophy changed in the face of the rapid development of AI - supported development and intelligent engineering tools?
Christian: Although AI can be a very useful helper for software architects, I believe that it cannot replace the technical decision - making process. Technical decisions must be guided by human skills and experiences.
In other words, I am firmly convinced that AI can only become a real productivity aid in software design when it is driven by humans, and not the other way around.
InfoQ: There are more and more AI - generated architecture suggestions (e.g., microservice division, optimization of dependency diagrams). Do you currently consider these tools reliable enough to become "architecture partners"? Or do they still remain at the level of "helpers"?
Christian: In my opinion, these tools help to propose possible solutions and discover new possibilities. But as I said before, they cannot replace the human decision - making process. In my view, these tools are and will always remain valuable "helpers" in the future.
InfoQ: Apiumhub works with many large enterprises on architecture transformations. Where do you see the greatest resistance in these projects - at the technical or cultural level? How can a sustainable "architecture culture" be built in an enterprise instead of just setting up an "architecture department"?
Christian: In an old article by Martin Fowler titled "Who Needs Architects?", he distinguished between two roles of traditional architects - the decision - maker and the leader, with the latter being the best way to "strengthen the team". This method helps to build a real software architecture culture in the team by working with developers, especially in the modeling phase.
However, I am firmly convinced that software architecture is a constantly evolving science in the broader field of engineering. I mean that anyone can become an architect, but it is not that easy. To become an effective architect, one must read a lot of scientific literature and continuously deepen one's understanding of this field.
Continuous Architecture and Architecture Observability
InfoQ: How can "observability" be integrated into system design so that architecture quality is not only on paper but visible and verifiable in real - time?
Christian: "Observability" usually refers to the quality characteristics of a system at runtime, which only represents a part of the overall quality of the system. Generally speaking, introducing architecture governance or design into a system means that all software properties must be continuously controlled. A good way to implement this process is to use techniques from "evolvable software architectures", where the most important concept is fitness functions.
InfoQ: In the face of complex ecosystems such as multiple AI agents, low - code components, and distributed systems, how would you recommend that architects build an "observability toolkit" at the architecture level? Which signals can indicate to the team that their architecture is becoming more and more "unobservable" and difficult to develop?
Christian: If we consider "observability" from the perspective of architecture governance, theoretically, an architectural error should be detectable through failed architectural unit tests. However, the signs of architecture degradation usually appear gradually, such as slower development speed, more errors, and runtime problems (e.g., performance problems), difficulties for the system to handle higher loads, and other similar symptoms.
InfoQ: In your book, you suggest connecting architecture quality characteristics (e.g., scalability, maintainability, modularity) with specific measurable indicators. How well is this idea accepted by teams in practice?
Christian: According to our experience, this actually depends on the specific situation of the team. Generally speaking, indicators should not necessarily be set as goals but should be carefully integrated into and spread in the team culture - that is the real starting point. Another important point is that the use of indicators should be based on real and recognized problems, that is, on problems that all team members feel and unanimously agree need to be focused on.
InfoQ: Which architecture indicators have you seen misused or misunderstood in practical projects? Can you give a specific example of how architecture indicators can help discover problems of "architecture degradation" in a system or make decisions about architecture redesign?
Christian: I think the most misused indicator is code test coverage. The problem is that this indicator tells us almost nothing about whether the test strategy is really effective - test effectiveness mainly depends on the quality of the design of the tested modules (e.g., classes, methods). Nevertheless, test coverage is still a useful signal when the number is very low, as this usually clearly indicates that team productivity is low or that there are problems in the development process.
InfoQ: With the entry of AI into the field of software analysis, is it possible that there will be "intelligent architecture monitoring" so that system indicators can not only be observed but also automatically optimized?
Christian: No, in my opinion, this still belongs to the category of "science fiction". Although AI can help in analyzing indicators and suggesting potential improvements, it cannot replace human judgment, nor the finely - tuned decision - making process in software architecture. Software architecture involves trade - offs, understanding of the business context, and prediction of future requirements, which are currently difficult to fully code and implement in an automated system.
The Golden Survival Law for Architects in the AI Era
InfoQ: Which characteristics do you think are the most important for a good architect in today's time? Is it analytical ability, leadership quality, empathy, or curiosity?
Christian: All three characteristics you mentioned, including analytical ability, leadership quality, and empathy, are important. Analytical ability is of course the key to understanding complex systems and making solid architecture decisions. But I think curiosity is also very valuable - it drives the continuous learning process, helps you learn new technologies and practices, and often leads to creative solutions that one would not find through analysis alone. In many ways, curiosity enables architects to grow and adapt in a constantly changing field.
InfoQ: Can you kindly recommend one or two books or resources that you consider currently or will be influential for the "thinking of architects" to our readers?
Christian: All well - known books on Domain - Driven Design (DDD) are very valuable. However, if you want to get deeper and up - to - date insights into software design, I recommend that you read the book "Balancing Coupling in Software Design" by Vladik Khononov.
InfoQ: If there is a "golden law for architects" that still holds in the intelligent era, what would you call it? Why?
Christian: The central "golden law" in our industry is: Remove the "I" from architecture. Architecture is a common vision, and one cannot make decisions based solely on one's own understanding of the business field. The actual architecture work should involve the entire team to ensure that all decisions are made collectively.