Articles
AI Delivery Programme
- AI agents (13)
- AI Coding Agents (1)
- AI delivery (88)
- AI Delivery Programme (12)
- AI Workflow Audit (2)
- Architecture Review (2)
- Laravel (1)
- MVP to Production (1)
-
Can AI code review take load off your human reviewers, and how do you test it first?
An AI reviewer can take part of the first pass on your pull requests. How much is a question for your own code: replay…
-
What working agreements should a team set for AI-assisted development?
Six working agreements for AI-assisted development (scope, size, tests, an author who can explain the change, unattended agents, owned instruction files), each held by…
-
How should a lead handle engineers who refuse AI coding tools?
Find out what each refusal is about, write down what every change must meet whatever tool produced it, watch for shared files and workload…
-
Should we rewrite a legacy system with AI, and how do we prove it still behaves the same?
Use AI to speed up a narrow, checkable migration, not as the reason to rewrite, and prove equivalence from the old system: characterisation tests,…
-
Can AI coding tools help on a legacy codebase nobody fully understands?
Where AI coding tools help on an old, many-handed codebase (reading, mapping, drafting notes, pinning behaviour with tests), where they make things worse, and…
-
Staff use personal AI accounts with company data: how do you get control without a ban?
A ban moves AI use to phones and home laptops. Inventory what people use and why, offer an approved business route that is easier…