Home

Database Systems Under Pressure

Enterprise databases are being reshaped by AI – not gradually, but structurally. This site examines what that means for the architecture, economics and safety of systems that cannot afford to be wrong.

Recent Writing

Two dark concrete slabs separated by a steel framework glowing teal and blue from within – the deliberately engineered layer that must be built between AI agents and enterprise databases

Something New Has to Sit Between Agents and Databases

Enterprise architecture depended on humans in ways nobody designed – and agentic AI is removing those dependencies. Something new has to sit between agents and databases, and it needs to be built deliberately.
Human silhouette defined by absence at the centre of a glowing enterprise data architecture, with circuit pathways and labelled system components built around the void

Enterprise Architecture Quietly Depended on Humans in Far More Places Than We Realised

Enterprise architecture was built around a human actor whose presence provided properties nobody designed and nobody priced. Agentic AI is the first technology to remove those properties simultaneously – and the absence is now visible across every domain.
A chrome rubber stamp making contact with a cloud of blue glowing particles, suggesting formal approval applied to something that cannot be fixed or fully characterised

Decades of Change Management. None of It Was Built for Agentic AI

Enterprise change management was built around software that behaves reproducibly – and that assumption is load-bearing in ways that only become visible when agentic AI removes it.
Data centre corridor with green server rack lights and a printed cost report in the foreground showing a single anomalous figure of $15,735.20 among otherwise uniform charges

You Won’t See Failure First. You’ll See Cost

Agentic AI introduces a failure mode that doesn’t announce itself. The first signal isn’t a red dashboard or a paged engineer – it’s a line in the cloud bill.

See all writing →

Blog Series

Databases in the Age of AI – a series exploring what AI agents mean for enterprise systems of record, transaction integrity, database performance and architecture

Databases in the Age of AI

AI agents are collapsing the layers between intent and the system of record. This series examines what that shift means for architecture, safety and economics – without pretending the answers are simple.
Database Performance In The Cloud – series index exploring the performance and architectural realities of running enterprise databases on public cloud infrastructure

Database Performance in the Cloud

Cloud infrastructure changes the rules for database performance in ways that aren’t always visible – until something goes wrong. A practical examination of latency, throughput and cost under real workloads.
Storage for DBAs – a series building the foundation of storage performance knowledge from first principles

Storage for DBAs

Flash changed everything about storage performance, but most DBAs inherited their mental models from spinning disk. This series builds the foundation from first principles.

Database architecture. No vendor affiliations. No easy answers.