Day 3 · 1:15 pm – 1:50 pm
Evaluating LLM Features Before Users Do
About this session
Test sets, rubrics and human review: a working evaluation setup for language model features.
When
Day 3 · 1:15 pm – 1:50 pm
Pick your stream
Three rooms, one afternoon
Three streams run side by side from 13:15 to 15:30. Move between rooms at the changeovers.
Stream
Stream D: AI in Practice
Evaluating, building and paying for AI features in real products. Runs in Main hall, Tuesday 8 December, 13:15 to 15:30.
See you at The Foundry
Diavola 2026 · 6–8 December · Bankside, London. Full passes from £549 + VAT, day passes from £249.