Your AI use-case list has thirty items and no owner
Long use-case lists feel like progress and function as a way of avoiding a decision. What replaces them is shorter, harder to write and actually gets built.
- AI strategy
- prioritisation
- use cases
Blog
Practical writing on getting AI into production: framing, ontology, data, agents, deployment and what happens after go-live.
Adoption is a budget line, not a hope
Every transformation plan has a technology budget and an adoption paragraph. That imbalance predicts the outcome more reliably than anything about the technology.
- change management
- adoption
- programmes
An agent is only as safe as the tools you gave it
Most attempts to make agents safe happen in the prompt. Almost all of the real safety comes from the tool layer, which is an engineering decision.
- AI agents
- security
- tool design
Design for the review, not for the answer
The metric that decides whether an AI feature saves anybody time is not accuracy. It is how long it takes a person to confirm or correct what it produced.
- UX
- AI interfaces
- review
Build the evaluation harness before the feature
Without a way to score changes, an AI codebase gets slower every month. Teams feel it around month three and usually blame the wrong thing.
- evaluation
- testing
- LLM
Most retrieval failures are definition failures
When search returns something technically related and practically wrong, the usual response is to tune the retriever. The usual cause is elsewhere.
- ontology
- retrieval
- master data
The answer was not wrong. The document was three months old
Freshness is the quietest failure in an AI system. Nothing errors, nothing alerts, and the answers are perfectly consistent with a version of reality that has moved on.
- data quality
- freshness
- pipelines
The AI policy people actually follow
Most AI policies are written to satisfy a risk register and are ignored within a month. The version that works is shorter, more permissive and considerably more useful.
- AI policy
- governance
- training
AI systems do not crash. They drift
Conventional monitoring watches for things that stop. The failure mode that matters here produces no error, no alert and perfectly formatted answers that are quietly worse.
- monitoring
- drift
- SLA
Go and watch a shift before you write a line of code
Field systems are rejected for reasons that never appear in a requirements document, and every one of them is visible within two hours of standing where the work happens.
- field deployment
- requirements
- adoption
When agents write the code, review becomes the bottleneck
Generation got ten times faster and review did not. Teams that measure only the first half accumulate under-reviewed code, and the debt arrives all at once.
- AI coding
- code review
- spec-driven
The night shift knows things nobody wrote down
Unattended production is rarely blocked by robotics. It is blocked by the judgement three experienced people carry, which exists nowhere a system can read.
- dark factory
- tacit knowledge
- context