Learn Core Concepts

The Get Started section gives you the working knowledge to build a function. The Learn tab goes a level deeper into the concepts that everything else in the platform is built on. Read these once and most of the rest of the docs becomes much easier to follow.

The core concepts

• Functions — what a function actually is, beyond “an endpoint” — model + samples + review queue + pipeline.
• Models — the AI systems inside a function; how Nyckel selects and manages them.
• Labels — the decision space of a classification function and how to design it well.
• Samples — the labeled examples that train private models.
• Confidence Scores — what the score means, how to read it, how it changes over time.

When to read them

You don’t need any of these to ship a working function — that’s the point of Get Started. Read the concept pages when:

• You’re making decisions about how to structure labels, samples, or your review workflow.
• You’re debugging accuracy and want to understand what’s happening under the hood.
• You’re integrating Nyckel into a larger system and need to reason about its behavior.

Next

• Improve Accuracy — practical guides for making your function better over time.
• Build Production Workflows — patterns for using Nyckel in a real application.