Hands-on techniques for evaluating AI systems.

Step-by-step evaluation methods, with worked examples and code. From confidence scoring to root cause analysis, the practical tools for measuring AI behaviour.

Dec 2025 8 min

Root Cause Analysis for LLM Behaviour

A step-by-step guide to diagnosing why an LLM behaves unexpectedly. Covers observability, categorisation, hypothesis testing, and reporting, with a template you can use immediately.

Read →

Nov 2025 7 min

Language Heuristics Part 1

Analysing text outputs for verbosity, language complexity and token performance. One of the simplest yet highly effective methods for checking LLM outputs.

Read →

Nov 2025 8 min

Confidence Scoring

Scoring how unsure the LLM is based on what you are asking it to do. Confidence is derived from logprobs and provides a direct insight into token-level certainty.

Read →

Oct 2025 10 min

Semantic Measurements

Scoring the meaning of natural language strings in a measurable way, so test outputs can be compared on what they mean rather than what they literally say.

Read →

Coming Soon

Upcoming 8 min

Human Evaluation

When and how to use human evaluators effectively. Covers inter-rater reliability, evaluation rubric design, and when human judgement is irreplaceable.

Preview →

Coming Soon

Upcoming 7 min

Perplexity

Understanding and measuring perplexity as a signal for LLM comprehension and prompt quality.

Preview →