TestingAI

Testing LLM Applications

Strategies for deterministic testing when your outputs are non-deterministic.

July 8, 20251 min read

Testing LLM applications requires a different mindset. When outputs are inherently non-deterministic, you need evaluation frameworks that measure quality on a spectrum rather than pass/fail assertions.

Evals Are the ProductWhy treating evaluations as a first-class product concern is the difference between shipping reliable LLM applications and guessing.4 min read

AITesting

Why I Still Reach for Boring Python FirstI default to stdlib and dataclasses until complexity is earned. Here is the heuristic I use to decide when a framework is worth the overhead.3 min read

PythonOpinionEngineering

Building a Design System from ScratchLessons learned creating a token-based design system with Tailwind v4.1 min read

Design SystemsTailwind

Stay in the loop

New posts on engineering, architecture, and what I'm building.