Articles
AI delivery
How teams deliver software with AI they can check: specifications, reviews, tests, records of what was checked, and the workflow around coding agents.
- 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)
-
Does AI coding make developers faster?
Results depend on the task and system around the tool, so compare delivery and stability on matched work instead of counting suggestions or generated…
-
How much does almost-right AI-generated code really cost?
Count the whole correction path: detecting the defect, rebuilding intent, changing code, rerunning checks, reviewing again, and recovering from escaped failures.
-
How do you keep AI-generated pull requests from overwhelming code review?
Measure arrival and completion by risk lane, reduce work in progress when review age rises, and improve evidence before adding more generation.
-
How do you stop an AI coding agent choosing the wrong files in a large repository?
Require a discovery pass that maps entry points, dependencies, tests, and local instructions before implementation begins.
-
How do you stop AI coding documentation from becoming outdated?
Tie each instruction to an owner and a change trigger, then test factual rules against the repository wherever a deterministic check is possible.
-
How do you stop new developers asking AI the same codebase questions every week?
Treat repeated questions as missing project memory; turn verified answers into maintained module notes, commands, decisions, or troubleshooting steps.