Section 158 · Chapter 21, Predictions for the Tokenized Product Future
The Last Engineers Standing
As AI takes over more creation work, the remaining human engineering leverage moves toward quality, safety, validation, and deciding what should be trusted.
What to do
- Define runnable checks that exercise RAG and last engineers standing.
- Set acceptable outcomes and blocker failures for RAG and last engineers standing before running the evaluation.
- Run representative cases for RAG and last engineers standing and preserve the failures that would change the decision.
Evidence to preserve
- Preserve the inputs, versions, configurations, raw outcomes, and results for RAG, last engineers standing needed to reproduce work on The Last Engineers Standing.
- Report results for RAG, last engineers standing by relevant slice, separate blocker failures from averages, state uncertainty and blind spots, and connect the result to a release decision.
Expert note
When the system matters, the enduring engineering role combines architecture, safety, measurement, incident learning, statistical thinking, security, human factors, and product judgment. AI can help produce artifacts, but humans still need to own the standards that decide whether those artifacts deserve power in their world.
Continue the conversation
Apply this to your context.
Save your product context once, then open a focused conversation that combines it with this concept.
Cite this page
Jason Arbon. "The Last Engineers Standing." Testing AI Knowledge Edition, section 158.
https://jarbon.ai/testing-ai/knowledge/ch158-last-engineers-standing.html