Sala Coding

La sala Coding è dedicata agli sviluppatori e tecnologi che vogliono scoprire come creare soluzioni AI. Approfondimenti su modelli, linguaggi, framework e best practice per rendere l’integrazione dell’AI accessibile e scalabile.
*Il programma è in via di definizione e potrebbe subire variazioni

Coding

Hosting della sala

Emanuele Arosio
Emanuele Arosio
Digital & SEO International Advisor - ASO Specialist - Seo & Digital Sales Leads
Emanuele Arosio
26 feb 11:40 - 12:20
40 min
In this presentation, we will introduce a recent proposal in AI that is called Retrieval-Augmented Generation (RAG). This technique is able to leverage the natural language capabilities of Large Language Model (LLM) to query a private database. You can literally ask questions about your data, without the need to fine-tune an LLM. We will build a simple RAG system using LangChain, an open source framework for writing applications for LLM. We will show with a few lines of Python that you can build a prototype system using a vector database like Elasticsearch.
26 feb 12:30 - 13:10
40 min
In today's financial and insurance sectors, AI-powered chatbots are transforming how institutions assist clients with their inquiries and problem-solving needs. These chatbots significantly enhance the client experience by simplifying the process of accessing information, learning about services, or resolving issues. Leveraging resources such as documentation, knowledge bases, and educational materials has never been more efficient. This presentation introduces a cutting-edge architecture for designing and deploying chatbots, with a focus on key components like data modeling, Retrieval Augmented Generation (RAG), and semantic caching. We will outline a standard technology stack for implementation and include a brief demonstration to showcase the potential impact on user interactions. Additionally, you will be invited to explore an open-source project that puts this architecture into practice, encouraging further engagement and innovation.
26 feb 14:10 - 14:50
40 min
26 feb 15:00 - 15:40
40 min
Gli agenti AI stanno ridefinendo il modo con cui automatizziamo i workflow aziendali, sbloccando immensi vantaggi di velocità, produttività, ed efficienza. Tuttavia, la progettazione di sistemi AI autonomi porta nuove sfide, dalla qualità e la sicurezza dei dati, all'affidabilità dei modelli di Generative AI moderni. Quali sono i pattern e le best practices? Come si costruisce un agente AI moderno, robusto e affidabile?
26 feb 15:50 - 16:30
40 min