JevMade field notes / Video guide

I Built an AI Project with Jev — The New System 1 Model by TypeSafe AI

DSwithBappy builds a Python customer-support triage demo using FastAPI. Jev selects a department, scores urgency, and checks refund and human-review conditions, while optional LangChain and Groq generate a suggested reply.

Original by DSwithBappyClassificationIntermediate50 min 32 sec Published 7 October 2026

Bookmark Keep this in your collection.

Leave a noteWhat would you try with this?

Only you can see your notes.

Before you press play

What you’ll find in the video

Jev acts as a System 1 decision model for fast, structured classification tasks, unlike traditional generative LLMs.
The application passes a state and typed questions to Jev to determine department, urgency, and refund probability.
Ordinary Python code handles the routing logic based on Jev's confidence scores, escalating uncertain cases to human review.

Worth knowing

Implementation and original video description verified; no code/API execution or full-video viewing. No tests in repository tree. Missing or malformed numeric values silently default to 0 and probability bounds are not validated. Demonstrated capabilities are attributed to this particular walkthrough.

All video guides →Prefer to read? Written guides →