Declan Gowran of IAS explores his role in the data engineering environment and how leaders are creating collaborative environments.
Senior data engineer IAS Declan Gowran’s journey into the world of data engineering came from a broad IT infrastructure and cloud background.
He told SiliconRepublic.com, “Early in my career, I worked extensively in enterprise infrastructure, virtualization and cloud deployments across multiple platforms, which exposed me to large and complex data management systems at scale. Over time, I became increasingly fascinated by the ways in which structured and unstructured data can drive decision-making and AI applications.
“That led me to roles at Optum and IAS where I could focus on building secure, unmanaged data platforms, integrating DevOps, MLOps and data management frameworks, as well as supporting AI business workloads. Basically, my approach was shaped by a combination of curiosity, technical challenges and opportunities to work at the intersection of cloud, data analytics.”
What is the current state of data engineering in Ireland?
Ireland’s data engineering landscape is vibrant and growing rapidly. With the strong presence of multinational technology companies and data-driven organizations, there is a growing demand for engineers who can not only manage cloud infrastructure but also design modern, unmanaged data platforms. Organizations are increasingly adopting cloud-native architectures, Kubernetes-based platforms and MLOps frameworks. There is also an emphasis on governance, compliance and data strategy, especially for companies that handle sensitive or regulated data.
What are the biggest challenges currently affecting the field of data engineering and how can they be solved?
Data management and trust at scale. As data powers AI and decision-making, ensuring quality, pedigree and secure access, while meeting regulations such as GDPR, is critical. This requires strong governance structures and centralized metadata to maintain consistency and control. Complexity in all distributed environments. Most organizations operate across multi-cloud and hybrid systems, making integration, configuration and planning difficult. The focus here is on simplifying architecture and using scalable, interoperable platforms to reduce fragmentation. Real-time performance measurement and AI-driven AI. There is a growing demand for low-latency data and reproducible AI pipelines. This means investing in a distributed, flexible and reliable infrastructure that can handle both cluster and real-time use cases. Overall, the solution isn’t just using tools, it’s combining these skills to define business outcomes, so data engineering draws more measurable value than technical skills alone.
What are you currently working on and what are its strengths?
I am currently leading the development of a secure, cost-optimized enterprise data platform at IAS, built on Databricks and Kubernetes. Designed to unify governance while enabling limited, automated access to data across the enterprise. In parallel, we develop AI gateways and services to support the safe deployment of LLM and AI workloads, ensuring that we can scale these capabilities responsibly. The power is twofold. Internally, greatly improving efficiency, teams can access trusted data quickly and easily test. In addition, it enables better products and results, from more effective ad campaigns to improved visibility and performance.
What goes into building a strong, cohesive team in data and engineering?
Creating data that works best with the engineering team requires balancing technical expertise with collaboration, culture and shared values. I strongly believe in investing in people and fostering a positive team atmosphere. It is not enough for team members to just understand technology, they also need to get along, communicate effectively and support each other. I specialize in coaching and development, clear communication, team alignment, cross-functional collaboration, breaking down silos, analytics, empowerment and autonomy. As well as providing developers with the right tools and frameworks to innovate while maintaining accountability. By putting people and culture first, we create an environment where trust, communication and collaboration are strong, allowing innovation and high performance to be natural outcomes.
How can leaders in dynamic spaces create productive and cohesive workplaces?
Leaders need to provide clarity, trust and organizational autonomy. This involves setting clear goals, fostering a culture of accountability and encouraging innovation without managing less. Using agile practices, automated workflows and transparent dashboards also help teams measure progress and stay on track. Equally important is supporting professional development, celebrating success and ensuring psychological safety so team members can collaborate freely and take calculated risks.
Do you have any predictions for how the data engineering space may evolve over the next nine months?
Over the next nine months, I anticipate several key trends shaping the landscape of data engineering. First, the widespread adoption of data management and management frameworks, especially for businesses managing AI and agency workloads, with a strong focus on data inventory, startup and integrity to know where data comes from, how it changes and why. Increased emphasis on data quality and protection from data poisoning, as organizations realize that “garbage in, garbage out” can compromise AI and agent model results. Greater adoption of native and serverless cloud architectures, enabling scalable, flexible and cost-effective platforms capable of supporting large AI workloads, agent processes and seamless communication across systems.
Advanced generation vector database expansion and connected pipelines, supporting advanced AI and agent use cases while ensuring embedded, information sources, and real-time data remain accurate, readable, and interactive. A strong focus on visibility, performance, and compliance, with distributed monitoring, automated validation and traceability becoming the standard for maintaining trust in both traditional data and AI results. Finally is the standardization of AI modeling and MLOps processes, allowing enterprises to standardize basic models, agent workloads and intelligent workflows while maintaining governance, reproducibility and operational reliability.
Don’t miss out on the information you need to succeed. Sign up for Daily BriefSilicon Republic’s digest of must-know sci-tech news.
