RAGAS
All frameworks
Evals and testing · by Ragas

RAGAS

Scores a RAG bot's answer and its retrieval separately, so you know which half to fix.

28 lessons3h 42m reading7 sections
01Evaluating a RAG app0/6 · 43m
RAGAS overview6m
Installation and setup6m
LLM evaluation6m
TechNest RAG app10m
Goldens7m
SingleTurnSample8m
Section quiz · 5 questionsNot taken
02Metrics without a judge0/2 · 12m
Exact match and string presence5m
ID-based context precision and recall7m
Section quiz · 3 questionsNot taken
03LLM as a judge0/3 · 25m
LLM as a judge9m
RAGAS metrics7m
Rate limits and cooldowns9m
Section quiz · 5 questionsNot taken
04The five RAG metrics0/6 · 54m
Faithfulness9m
Answer relevancy9m
Context precision9m
Context recall10m
Answer correctness10m
Factual correctness7m
Section quiz · 5 questionsNot taken
05Your own metrics0/2 · 13m
Custom metrics6m
DiscreteMetric7m
Section quiz · 3 questionsNot taken
06Running the evaluation0/6 · 44m
Datasets6m
Experiments8m
Scoring an experiment10m
Evaluation results7m
evaluate() and EvaluationDataset7m
Collections and legacy metrics6m
Section quiz · 5 questionsNot taken
07Shipping the evaluation0/3 · 31m
Evals in CI8m
Integrations8m
Project: TechNest support bot evaluation15m
Section quiz · 4 questionsNot taken