
The dangerous part was not the classifier. The label file was, sitting there looking convenient. A saved train_y.npy artifact existed, and for this baseline it was a trap. It contained magnitude-filte
AI engineer. LLM agents, RAG, Azure AI Foundry, Python/TypeScript. Navy veteran. Builder of Scenematic, Pramaana, and production multi-agent systems. I write about what I actually ship.
Sr. AI Engineer & Architect. Building deterministic, context-aware AI for enterprise production.
Markdown supported · ⌘↵ to post
No responses yet.