JevMade

Sign in
← Back to videos

JevMade field notes / Video guide

Why Jev's creator says classifiers and small code checks beat giant prompts

Matt Turck of the MAD Podcast interviews Diogo Almeida, chief executive of TypeSafe, about System One models, how email automation can be split into measurable classifiers and code, and why he distrusts benchmarks and chat-style training.

Original by The MAD Podcast with Matt TurckGetting startedOverview67 min 0 sec Published

Before you press play

What you’ll find in the video

  1. The interview argues that splitting automation into small classifiers plus ordinary code is more reliable than one giant prompt.
  2. Probabilities and thresholds are presented as the way developers keep control over when software acts.
  3. Benchmark scores are criticised in the interview, so test candidate models on your own data.
Worth knowing

An interview with the maker's own chief executive, so claims about Jev, copycats and benchmarks are his opinion and were not tested by JevMade. The summary rests on the published chapter list and show notes, not a full viewing. The duration is the roughly 67 minutes the podcast listing gives, so it may be a minute out.

Why Jev's creator says classifiers and small code checks beat giant prompts

Your note

Only you can see this. Clear the text and save to remove it.

Keep what you save

Sign in to keep bookmarks and private notes on every device. We email you a link; there is no password, and the same link creates your account.