Evals.cz

Evaluating AI applications, not just models.

Evals.cz is a Prague-based meetup for people who build or explore AI-powered products. Model benchmarks on leaderboards are interesting, but how do you evaluate a specific RAG pipeline in production?

Next meetup: January 14 at Filevine — agenda and registration coming soon

Subscribe to the event calendar on Luma

What we cover

RAG System Quality

Evaluating retrieval quality and end-to-end RAG pipelines in production

Production Regressions

How to detect and fix issues in deployed AI systems

Human-in-the-loop

Evaluation approaches combining human expertise with automation

Development Lifecycle

Integrating evaluations into development and CI/CD pipelines

Who is it for

ML/AI Engineers

You deploy and monitor models in production

Backend Developers

You integrate LLMs or RAG into existing products

Data Scientists

You measure and improve AI system performance

Product Managers

You need to understand AI feature quality

AI Researchers

You study AI reliability and safety

Beginners Welcome

Practical interest and curiosity are welcome

Meetup format

Venue Filevine (Prague)
Format 1-3 short talks + discussion
Language Primarily English
Atmosphere Informal, friendly, practical
"What broke in production and how we fixed it" — no sales pitches

Supported by

Education

Intro

AI Evals 101

Two-hour online session on evaluation fundamentals. Concepts, live demo and Q&A.

2 hours liveOnline3,000 CZK
More info
Methodology

What you will learn

  • Error analysis & failure taxonomy
  • Assertion-based evaluation
  • LLM-as-Judge framework
  • RAG-specific metrics
  • Agent & tool-use evaluation
  • Human evaluation & annotation design
  • Integrating evaluations into CI/CD