Data costs are where cloud costs were a decade ago. The maturity curve from ignorance to continuous optimisation, and where most data teams actually sit on it.

Data teams have decades of intuition about CPU compute. GPU economics break most of it. The utilisation problem, the build/rent question, and what actually pays back.

Cloud egress is the slow tax. It doesn’t show up in the architecture diagram. It doesn’t show up in the storage bill. It shows up as a line item once…

In most orgs, query cost is not something we bake into the design phase (mainly because we don’t know…duh). But it’s then subsequently forgotten all the way through the production…

“A petabyte costs about $23 a month in S3.” You may hear this (you may not if you have a social life). It’s not really true, in the sense that…

LLM hallucinations on unstructured data have been known for the last few years. The chatbot makes up a citation and the summariser invents a quote. The RAG system produces a…

Evaluating an analytics agent (or team of agents) is WAY harder than evaluating a chatbot, harder than evaluating a search system, and considerably harder than evaluating a regular ETL pipeline.…

ETL has been the same conceptual thing for 50 years. Tool use is what it becomes when the orchestrator is an LLM. What changes, what survives, and why MCP matters.

Embarrassing legacy BI vocabulary five years ago, back at the centre of the data stack today. The dbt / warehouse / OSI convergence.

Pointing your agent at the Confluence docs isn't enough. What 'agent-ready metadata' actually means, and why most enterprises fail the metadata-only test.

Text-to-SQL has been almost-working for five years. Why the schema problem and the semantic gap keep biting, and what production deployments actually look like.

PII discovery is the unglamorous problem every governance vendor pitches as solved. It isn't. Why pattern matching, ML, and the shift-left approach all fall short.