Mastering RAG: Overcoming Key Challenges and Handling Multi-modal Data

FREE EVENT, ONLINE, PLEASE REGISTER ON EVENTBRITE

Join us for an in-depth look at Retrieval-Augmented Generation (RAG) and multi-modal data integration, with two expert led sessions designed for tech leads, developers and data experts.

See real prototypes in action as you learn to build efficient RAG systems capable of processing diverse data – text, images, tables and video – powered by generative AI.

Learn effective strategies to improve RAGs performance, tackling toughest challenges like accuracy, scalability and real-time data integration in enterprise applications.

Agenda

11:00 – 11:05 I Welcome, KI Park

11:05 – 11:25 I RAG systems for Multi-modal, Data Taras Hnot, Data Science Principal, Consultancy Lead at SoftServe

The talk will focus on the analysis of how multimodal data retrieval systems can be efficiently built. Imagine vast amounts of documents containing not only text but also valuable information encoded in schemas, images, and complex tables, along with videos that provide instructions and visual explanations. The obvious question that arises is: how can we process this data efficiently and build a pipeline capable of understanding all of it? During the session, we will cover the architecture of such a solution and examine a real, implemented prototype that demonstrates the tremendous opportunities now available through multimodal Generative AI models.

11:25 – 11:45 I Overcoming Limitations in Retrieval-Augmented Generation (RAG), Asaad Almutareb, Founder, Artiquare

Retrieval-Augmented Generation (RAG) has been transformative in combining generative AI with relevant, domain-specific information, yet it faces distinct limitations in accuracy, scalability, and real-time data integration. This talk delves into RAG’s primary challenges, such as dependency on retrieval quality, handling evolving data, and response latency. We’ll explore advanced strategies to counter these issues, including optimized retrieval algorithms, hybrid architectures, and fine-tuning models for specific applications. Attendees will gain insights into practical solutions for elevating RAG’s effectiveness in real-world enterprise environments.

11:45 – 11:50 I Summary

PLEASE REGISTER ON EVENTBRITE

Speaker bios

Taras Hnot, Principal AI Consultancy Lead at SoftServe, has more than 10 years of experience in high-tech projects, including technologies such as Advanced Data Analytics, Deep Neural Networks, NLP, Reinforcement Learning, Generative AI, and others. Taras's mission is to lead and develop a team of top-notch AI experts and elevate the role of AI and Gen AI as one of the primary business development directions for the company.

Asaad Almutareb, the Founder of Artiquare, an AI-driven startup pioneering intelligent, knowledge-driven automation through cutting-edge technologies like AI Agents and Retrieval-Augmented Generation (RAG). With 13+ years of experience leading software development in production automation, big data and predictive analytics for the automotive sector, he is passionate about leveraging open-source LLMs to democratize AI and redefine innovation.

Who should join us?
Tech savvy professionals who want to stay at the forefront of RAG systems

• Developers/ software engineers,
• Data scientists, data engineers
• Technical researchers,
• Tech leads, and engineering managers
• Tech savvy founders
• Enterprise AI and machine learning specialists
• Product managers in AI/ML and data products
• Researchers and academics in AI and machine learning