Short, specific pieces on certification evidence, traceability, and review practice — the problems that actually show up in DO-178C, ISO 26262, EN 50128, and IEC 62304 programs, and what fixes them.
Written by practitioners who have prepared and defended certification evidence — not repackaged standards summaries.
You can spend $7.5 billion and still not get certified. Three programmes that prove it — SpaceJet, the KC-46 tanker, and Eclipse Aviation — and why the overrun always begins at requirement review.
A traceability matrix in a spreadsheet is a photograph of a relationship that keeps changing. Why matrices decay silently, what bidirectional traceability really demands, and a side-by-side of what changes when trace data becomes a view rather than a copy.
Most audit findings are not about bad engineering. They are about evidence that could not speak for itself. Five recurring findings, written the way an auditor would record them — and the single root cause behind all of them.
A design change moved through engineering without one record connecting it to its updated impact analysis. What the 737 MAX MCAS investigations found about change control — and what it takes to keep that record whole.
The code that destroyed Ariane 5 Flight 501 was correct against its requirement. Nobody re-engineered that requirement for a new vehicle — because the module was reused, not reviewed.
The F-35 program tested constantly for two decades and still paid billions in retrofit costs. GAO's own term for why: concurrency — production running ahead of the verification evidence meant to justify it.
When investigators asked whether the door plug bolts on Alaska Airlines Flight 1282 had been reinstalled, no record could answer them. What that gap reveals about treating audit readiness as a standing state.