In my previous post , we explored the big changes required in architecture to make LLMs successful. As those who are familiar with me will know that quality and testing are vital aspects of software architecture to me. So, I feel like I wouldn’t be able to explore the topic of software architecture without aspects of testing and quality. Testing and QA Must Be Built Into Architecture In classical software engineering – at least at a fundamental code level - testing often tends to follow a...

As I’ve already explored in my previous articles,  LLM-powered applications are having a big impact on the way we think about software development. With this rapid acceleration into AI adoption, many companies and teams are discovering that building applications powered by Large Language Models (LLMs) feels nothing like  building traditional software. The patterns are different, the risks are different, and so are the architectural decisions. LLMs introduce new forms of complexity,...

Large Language Models (LLMs) are only as good as the data that shapes them. Whether you’re fine-tuning a model for domain-specific use or building an LLM-powered application from scratch, data readiness is the single biggest factor influencing performance, reliability, and ethical outcomes. As the saying goes, “garbage in, garbage out”  - but with LLMs, the cost of poor data goes beyond technical glitches; it can lead to bias, misinformation, and loss of trust. In this post, we’ll explore...