My recent posts.

21 articles

LLMs feel smart, but don't actually understand anything

LLMs can sound like they understand you, but under the hood they're predicting the next token, not forming meaning. That's why they can produce beautifully structured nonsense, so the right approach is verification and constraints, not trust.

11 min read

Fueled by .NET, curiosity, and matcha.

© 2026 Lindsey Broos