Systems, Not Silos

AI can help engineers move faster, but speed alone is not software engineering. The real test is whether the system remains understandable, secure, testable, maintainable, and useful after the first version is shipped. I have been using AI tools more and more in my day-to-day engineering work, including ChatGPT, Claude, GitHub Copilot, Cursor, Codex, and Gemini. They are genuinely useful for…

Building software for government, transport, infrastructure, and essential services is not just about writing code. It is about turning complex rules, operational knowledge, legislation, policies, data, and stakeholder requirements into systems that people can actually use. That is why I find the intersection of full stack development, AI platforms, GraphRAG, knowledge graphs, and public-sector…

Senior systems engineering is one of those disciplines that rarely gets attention when everything is working, but quickly becomes critical when something breaks.The role is not just about keeping servers online.It is about understanding how an organisation really operates through its infrastructure, identity systems, databases, integrations, cloud services, monitoring, backups, support processes,…

Digital Leadership Is About Outcomes, Not Just Modernisation In many organisations, digital transformation is still treated as a technology program. New platforms. New dashboards. New integrations. New cloud environments. New tools. All of those things matter, but they are not the real measure of success. The real test is whether technology helps an organisation make better decisions, deliver…

Large ICT programs rarely fail because one spreadsheet was wrong. They usually fail because finance, delivery, architecture, risk, and operational reality drift apart. That is why program finance in digital transformation is not just about reporting numbers after the fact. It is about building decision confidence while the work is still moving. In complex ICT programs, finance needs to…

One of the most important lessons I have learned across software engineering, technical leadership, logistics platforms, payments, eCommerce, operational systems, and enterprise integrations is that the best systems rarely begin with code. They begin with understanding. Understanding the business process. Understanding the user. Understanding the data. Understanding the legacy system that…

In many organisations, the role of IT leadership has changed significantly. It is no longer enough to simply “keep the systems running,” but it is also not enough to sit too far above the work and only talk in terms of frameworks, roadmaps, and abstract strategy. The best technology leaders, especially in operational environments, need to sit somewhere in the middle. Close enough to the…

Software architecture should not live in a document that gets handed over and forgotten. The best architecture work I have seen, and the kind I enjoy doing most, sits close enough to delivery to remain useful. That does not mean architects should micromanage engineers or turn every implementation detail into a design committee. It means the architect remains accountable for the integrity of the…

There is a big difference between using AI to generate code and designing AI-enabled engineering systems that can safely ship software. That difference is evidence. A single AI assistant can help write a function, explain an error, create a test, or refactor a component. That is useful, but it is not enough for production engineering. Once AI becomes part of a team workflow, the question is no…

One line in a recent Principal AI Engineer job advertisement stood out to me:“Ideally you’re a software engineer first, ML engineer second.”That single statement captures one of the biggest shifts happening in AI engineering right now.For years, many organisations treated AI as a research problem.Build a model. Run experiments. Create a proof of concept. Present some impressive metrics.Then…

One thing I have learned over the years is that senior engineering is rarely about writing the “perfect” code. It is about making good decisions when systems become unstable, integrations fail, customers are affected, and the business waits for answers. That is why so many technical interviews today are moving away from textbook questions and focusing more on realistic engineering…

Over the years, one thing became very clear to me: The hardest engineering problems are usually not about code. They are about operational trust. In industries like logistics, transport, mining, warehousing, aviation, and industrial operations, the real challenge is often this: “How do we make systems reliable enough that people trust operational decisions based on them?” A delayed telemetry…

Over the past few weeks, I have been reflecting on something uncomfortable happening across the technology hiring market. Not AI itself.Not ATS platforms alone.Not even recruiters. The real problem is what happens when everybody stops trusting everybody else. I recently applied for a senior leadership role at a well-known technology company. The recruiter came back saying my ATS score was…

One thing has become increasingly clear to me recently: The companies getting real value from AI are usually not the ones chasing hype. They are the ones fixing operational friction. A lot of current AI discussions still revolve around:• prompts• models• agents• frameworks• tooling stacks But inside real businesses, especially across logistics, SaaS, enterprise systems, and operational…