Designing Your Own Generative AI Assistant: From Concept to Creation

Generative AI assistants are chatbots which are based on large language models (LLM) which are built using transformative neural networks. Those AI assistants are playing an increasing role in society thanks to their good ability to generate text, images, sounds and human-like videos. They can be used in various areas such as: customer service, personal assistants, education, healthcare, accessibility, content creation, entertainment and much more.

However, while AI assistants offer many benefits, their usage in a specific context needs to understand how it is possible to tailor them to our specific needs.

This workshop presents the basis of  Generative AI assistants and introduces with practical exercises or demonstrations how to design a specialized AI assistant.

The main parts of the training are:

• Quick introduction to the fields of artificial intelligence, deep learning, Large language models (LLM) and self-attention networks (Transformers).
• Design approaches to custom AI assistants:
• • prompt engineering
  • personas / personas orchestration
  • retrieval augmented generation (RAG)
  • fine tuning
• Demonstration and practice of the design approaches to custom AI assistants.

• Understanding of the main aspects to consider for the definition, design and implementation of a generative AI ChatBot
• Apply basic prompt engineering, persona engineering and retrieval-augmented generation technics to create a custom generative AI ChatBot assistant

Lecture  (slides presentation, theoretical instruction ...) 
Practical exercices  (problem-solving, coding exercises, modeling exercices, quiz...)