LLM

Vibe coding needs testing, and the importance of "Intent" resolution in retrieval.

Feeling the vibe, when fat gets thrown out for thin.

Digging into tool-calling, code-calling, and courtrooms.

I discuss connecting to external LLMs, adaptors, routing & redaction.

For those curious about how this thing actually runs in the real world, we briefly step away from LLMs and…

With this post, we explain how to see inside our LLM discussions, when things go wrong.

Deeper in the machinery - calling tools, and avoiding the apocalypse.

Where Failure Becomes… Well, Mostly More Failure (But Also Progress)

Normalisation is the unglamorous but essential work of turning messy human questions into clean, predictable input that an LLM can…