Engineering delivery · automotive validation · digital products
Engineering & applied AI notes
Practical methods for making technical work easier to review. These articles describe approaches and worked examples, rather than claims of measured deployments.
01 / Engineering note · Automotive / Applied AI
CANalyzer 15 Assistant: a reusable AI skill for evidence-based bench validation
A reusable AI skill for CANalyzer 15 CAPL and trace workflows, with explicit timing criteria, evidence limits and a proposed offline CSV checker.
Read the article02 / Engineering note · Validation practice
Make a validation finding reviewable: context, expectation, evidence
A practical structure for validation findings: prerequisites, expected behaviour, observed evidence and a clear pass, fail or blocked conclusion.
Read the article03 / Applied AI note · Product evaluation
Define acceptance criteria before adding AI to a workflow
A worked approach to scoping an AI assistant: representative examples, deterministic checks, grading rubrics, human review and release evidence.
Read the article