[FREE] Fast and cheap AI classification using Jev model

Mandatory free registration:

https://stacja.it/produkt/bezplatnie-szybka-i-tania-klasyfikacja-ai-z-wykorzystaniem-modelu-jev/
Evening workshop at 17:00-21:00 (4 hours)

What will you learn?

During the workshop, you will get to know the Jev model, which is a fast and cheap AI classifier. Unlike large language models, Jev does not generate text, but is able to perform a classification based on it. You will learn the types of classifications that Jev gives access to, you will see in practice how this model works. You will also learn the difference between a Jev and an LLM.

Who is this workshop for?

The workshop is designed for people who know Python, have at least basic experience with LLM and want to learn the possibilities of the new Jev model.

Prerequisites

Knowledge of Python and basic experience with LLMs via API is required. It will be useful to know the tool JupyterNotebook / JupyterLab or another file editor. ipynb, because we will use it during the workshop.

The meeting will be led by Patryk Palej:

Since 2018, he has been involved in data analysis, machine learning and programming in Python. In the meantime, he completed studies in the field of Energy, during which he was engaged in mathematical modeling of heat and mass transport processes. However, he quickly switched to IT and as a data scientist worked on projects for such sectors as finance, telecommunications and media. He currently works at Ailleron on solutions for the banking sector based on artificial intelligence. In Sages he conducts trainings, workshops and classes on issues on the border of Python and data science.

Program

1. Generative model and decision model – when AI should respond and when to decide.
2. Build a Jev query: state, questions and response types (choice, score, noul).
3. Probability and confidence – automatic decision thresholds and pass for manual verification.
4. Build your own classifier and assess its effectiveness on test data (spam, sentiment, service requests).
5. Border cases, deterministic and effective for Polish and English.
6. Jev vs. LLM vs. Classic ML – models compare cost, uptime and accuracy.

UWAGA

Places are limited! The organizer reserves the right to cancel the event in case of failure to register the minimum number of participants.

Mandatory free registration:

https://stacja.it/produkt/bezplatnie-szybka-i-tania-klasyfikacja-ai-z-wykorzystaniem-modelu-jev/