A company’s technical architecture is often at the core of everything it does, maintaining it is therefore a key priority.
Technical architecture, from an organisational point of view, is the software, hardware and networks that sit at the centre of a company and is largely responsible for both big and small objectives. That is to say, this form of architecture is crucial to a business and is essentially its foundation.
SiliconRepublic.com spoke with various leaders in the STEM space about the importance of technical architecture, the challenges of maintaining it and how pipelines can be developed to support its development, as well as what it takes to conceptualise and implement future tech.
“I believe that one of the biggest challenges is not designing the future architecture, but transitioning to it,” said Natasha Bornhorst, a senior software engineer at Rent the Runway.
“This is particularly challenging when you’re re-architecting a cornerstone monolithic service into more manageable and scalable microservices. It’s tempting to want to simply build an entirely new system on the side and switch over in one large cutover, but this can introduce significant operational and business risks.
“Instead, we have found benefit in evolving the architecture piece by piece. We isolate individual components, pull them out and re-architect them, gradually cutting over to the new solution while keeping the existing system operational.”
This view was echoed by Mark Greville, the vice-president of architecture at Workhuman, who compared attempts to keep pace with AI and architectural demands to “juggling chainsaws on a single ski while being pulled by a speedboat”.
He explained: “Vibe coding has increased development velocity, but also opened new attack vectors. There is no standard AI SDLC [AI software development life cycle] and the right model today may be outdated in a month.”
He suggested adopting five main policies to be implemented in line with the transformation of architecture.
These included: an adherence to strong governance; the creation of a centralised AI group to enable a standardised system across teams; built-in infosec to identity and address threats from day one; written context in machine readable format for every step taken; and significant human oversight so as to never replace accountability.
Man versus machine
For BearingPoint’s Declan Leonard, the director of technology architecture at BearingPoint Ireland, the biggest challenge is tied into the idea that advancements, however innovative, are always subject to a review and strict governance – and they can always be improved upon.
He said: “At its core, architecture is about making informed design decisions that balance business objectives, legacy context, security, resilience, scalability, cost, usability, regulatory obligations and long-term maintainability. However, as organisations become increasingly digital and interconnected, the number and complexity of these considerations continue to grow.
“For organisations delivering critical services, architecture needs to be intentional from the outset and continuously governed throughout the life cycle. Security, in particular, must be treated as a first-class design consideration, not something added at the end.
“It needs to be embedded across the entire secure software development life cycle, from requirements and threat modelling through build, test, deployment, monitoring and continuous improvement.”
But, for technical architecture to be maintained and improved to a standard that supports a company’s evolution, a strong, highly skilled and consistent talent pipeline is critical, he said.
Top talent
“A consistent talent pipeline is important as it creates an environment of innovation and fresh thinking for the architecture group and growth for the individuals themselves,” said Eoin Honer, the managing director for technology and transformation at Deloitte Ireland.
“It can be the difference between architecture that evolves and innovates and architecture that perhaps doesn’t. Without a steady flow of skilled people, senior architects end up being overloaded or worse still, a single source of knowledge which can have negative outcomes.
He added: “Architecture needs both seasoned professionals who understand the domain deeply and newer voices asking, ‘why do we do it this way?’. Invest in relevant training and pathways to prepare your people to be the architects of the future.”
“It is hugely important”, agreed Gerard McMahon, the vice-president and engineering leader for software delivery at Fidelity Investments Ireland. He believes that the best technical architecture comes from the people who understand how systems are actually built, how they are deployed, supported, secured and how they evolve over time.
McMahon said: “AI will make architectural knowledge more accessible to engineers in the flow of work, but that does not remove the need for sound technical judgement. If anything, it raises the bar.
“We will need people who can work effectively with AI, challenge its suggestions, understand the trade-offs and keep architecture grounded in real delivery. A strong talent pipeline should therefore develop engineers who can combine hands-on experience, systems thinking and AI-enabled governance, rather than treating architecture as a separate review activity.”
“None of us starts out as an architect”, noted Greville, who further explained that we all learn from those around us and by developing a strong talent pipeline, organisations welcome new ideas, challenge outdated ways of thinking and create a stronger business model based on a range of backgrounds and experiences.
Forge your own path
Bornhorst finds that while it is important to take into account other people’s views, it is imperative that you learn how to solve problems independently. She said one of the biggest issues that held her back in the early days, as a fresh-out-of-college engineer, was the belief that there is only one right way to do things.
“I leaned too heavily on those above me, assuming they had the correct answer. But solving problems in a codebase is not the same as taking an exam in college. There is rarely one right architectural solution, there can be many ways to solve a problem, each with its own pros and cons,” she said.
“Don’t be afraid to go into the details, ask questions, read the code and do the research yourself. This can help you become an expert in an area of the codebase that has no experts. Taking ownership of a complex or poorly understood area is a great way to build deep technical knowledge and develop the confidence needed to make good architectural decisions.”
Honer too believes in the value of building a good reputation early on, particularly a strong technical foundation that forms the basis for credibility.
He said: “From there extend your curiosity beyond your immediate area and really think about how your work connects with everything else and ask yourself what purpose is being served in the broader context. Understanding how each piece fits in helping the organisation achieve its goals, that’s when architecture thinking starts to form.”
“Be curious, ask good questions, document your thinking and build a reputation as someone who can connect technical detail with practical business outcomes,” concluded McMahon.
Don’t miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic’s digest of need-to-know sci-tech news.












You must be logged in to post a comment Login