Скачать с ютуб Israeli Engineering Leaders Forum 2/4/25: Data Platforms with LLMs (Hebrew) в хорошем качестве

Israeli Engineering Leaders Forum 2/4/25: Data Platforms with LLMs (Hebrew)

​As LLMs become increasingly central to AI strategies, engineering leaders face the challenge of making them reliable, efficient, and truly data-driven. The key lies in a strong data platform—one that not only manages data efficiently but also enhances its quality for AI applications. This webinar will address the pressing need for a self-service data platform to simplify data preparation, improve LLM reliability, and scale advanced search systems effectively. We'll discuss how a well-structured data platform can streamline data enrichment, support practical testing approaches for LLMs in production, and enable the integration of generative AI with hybrid search capabilities. This is about moving beyond the hype to practical, actionable strategies that engineering managers can implement to make AI work better. ​On the Agenda: Opening words // Tamir Tausi, Head of Sales & Account at Tikal. Powering LLMs: Why a Robust Data Platform is Essential // Chaim Turkel, Group Leader & Data Architect at Tikal ​Many view Large Language Models (LLMs) as purely a distributed application layer, rather than an integral part of a data platform. However, at the core of every LLM lies one fundamental truth—your answers are only as good as your data. In this talk, we’ll explore how a self-service data platform can supercharge LLM applications through data enrichment and preparation. From leveraging structured pipelines to using LLMs themselves for data enhancement, we’ll dive into practical strategies to ensure your AI is not just intelligent, but truly data-driven. Enhance AI Applications Reliability Using Your Data Platform // Shani Cohen, Senior Backend & Data Developer at Tikal ​While the operational requirements for creating LLM applications are similar to those of traditional software applications, there are unique challenges with testing LLMs that require a distinct approach. This talk will provide an overview of a use case demonstrating the required updates for your engineering team to evaluate LLM in production successfully. ​Building Conversational Search Engines with Generative AI & Hybrid Search // Shahar Satamkar, Senior Algorithm Engineer, Fiverr ​Writing effective search queries is difficult, especially when users have limited input. Traditional search engines rely on keywords, but Generative AI and semantic search allow users to express themselves more freely and receive better results. At Fiverr, we’ve combined these technologies in our Fiverr NEO chat-based search engine to deliver more relevant results. In this session, I'll share how we built a hybrid search system that blends keyword and semantic search, using Generative AI to improve query understanding and continuously optimize results with real-time feedback. We'll code together a chat-based search engine from scratch, see how to evaluate its results' relevance, monitor it in production and scale it up using VectorDBs and traditional information retrieval approaches. Explore more Tikal Events: Meetup group: https://tkl.to/tikal-meetup-israeli-t... Lu.ma: https://tkl.to/tikal-lu ma Dive into insights on our Tech Blog: https://tkl.to/tikal-medium-israeli-t... Follow us on LinkedIn: https://tkl.to/tikal-linkedin Tune into Tech Radar Voice Podcast: https://tkl.to/tikal-tech-radar-voice...