Brilliant.org, but for LLM evals

Learn LLM evaluation
by actually doing it

Tap-through lessons that teach RAG evals, agent evaluation, and production monitoring the way Brilliant teaches physics — through manipulation, not memorization.

What a lesson feels like

Not a tutorial.
An experience.

One idea per screen
Zero walls of text
Working code you can run today
Real prod incidents to diagnose

What’s the faithfulness score?

Faithfulness

Partially Faithful

✓ In the doc

✗ Made up

Six tracks · 24 lessons

The complete map of LLM eval

Every lesson teaches architecture, implementation, production monitoring, and how to know when you're ready to ship.

🔍

RAG Evaluation

Live

Faithfulness, context precision, answer relevance, recall, pipeline scoring

RAGASDeepEvalTruLens

lessons

💬

LLM & Chatbot Eval

Live

Hallucination at scale, response quality, safety & toxicity, multi-turn eval

DeepEvalGuardrails AIPromptfoo

lessons

🤖

Agent Evaluation

Live

Task completion, tool-use correctness, trajectory efficiency, multi-agent systems

DeepEvalLangSmithAgentBench

lessons

🧠

Fine-Tuned Models

Live

Behavioral regression testing, benchmarking, alignment verification

EvalsBraintrustlm-evaluation-harness

lessons

📊

Production Monitoring

Live

Observability, online vs offline eval, drift detection, SLOs, A/B testing

LangSmithPhoenixBraintrust

lessons

⚙️

Eval Architecture

Live

Eval harness design, CI/CD gates, LLM-as-judge calibration

GitHub ActionsPromptfooWeave

lessons

Meet the token characters

Faithful

Retriever

Judge

Hallucinator

Start with one lesson.
7 minutes. No setup.

Did it actually use the document? ↓

Learn LLM evaluationby actually doing it

Not a tutorial.An experience.

The complete map of LLM eval

Meet the token characters

Start with one lesson.7 minutes. No setup.

Learn LLM evaluation
by actually doing it

Not a tutorial.
An experience.

Start with one lesson.
7 minutes. No setup.