You have decided on AI. Now the choices that decide success.
Most organisations use AI; few get measurable value from it. Five implementation choices make the difference, and the right answer depends on context.
Read article →Insights
Articles on AI systems, IoT systems, software solutions and engineering, drawn from real projects and backed by sources.
AI systems
Notes from building AI agents and pipelines that run in real work, not in demos.
1 article →IoT systems
Notes from building sensor hardware, firmware and data pipelines for occupied buildings.
1 article →Software solutions
Notes from designing and running multi-tenant platforms in continuous operation.
1 article →Engineering and design
Notes on systems engineering, from first concept to designs that get built.
1 article →All articles
What large language models change for IoT
IoT has long produced more data than organisations could use. Large language models act on the layer that blocked value: meaning.
Read article →AI makes coding easier. It also creates new requirements.
Writing code has become faster. The bottleneck has moved to specification, review, testing, security and accountability.
Read article →Engineering in the age of AI
AI speeds up documents, checks and design options. The engineer’s role shifts from producing to specifying, verifying and deciding.
Read article →Working on something in these fields? Write to me.