Skip to main content
Xalicon

Insights

Engineering writing, not marketing copy

Notes from the work: how we evaluate AI features, choose tenancy models, migrate legacy systems and keep performance from regressing.

These articles are in editorial review

Everything here is written in-house and currently marked as draft while it goes through technical and editorial review. The reasoning is ours and we stand behind it; the draft label simply means it has not yet been through final review.

FeaturedAI EngineeringDraft

Build the evaluation set before you write the prompt

Most AI features stall because nobody can tell whether a change made things better. A small, honest test set fixes that faster than any prompt technique.

Xalicon Engineering7 min read

Read the article

Get new articles by email

Roughly monthly. Engineering writing only, unsubscribe in one click.

Start a conversation

Want to talk about any of this?

These articles describe how we work. If something here matches a problem you have, we are happy to go deeper.

Prefer email? contact@xalicon.co