Shaikh Quader – IBM
Abstract
Large language models (LLMs) have transformed AI, enabling a new class of applications known as language apps. Enterprises are increasingly adopting these apps for tasks such as chatbots, document summarization, question-answering, and information retrieval, driving renewed interest in AI. A key component of language apps is the generation of vector embeddings using language models. This process converts unstructured content, such as PDFs, into AI-readable representations. These embeddings power vector similarity search, which delivers more precise semantic retrieval than traditional search methods. Vectors require efficient, cost-effective, and secure storage, along with optimized search techniques. To address these needs, Db2 is introducing its first phase of vector support, featuring native vector type, vector similarity search, and a range of vector operations. In this presentation, the speaker will explore Db2’s upcoming vector capabilities and how they can be leveraged to build modern AI applications with Db2.
Speaker Bio
Shaikh Quader is the AI architect at IBM Db2. His role involves developing product features to enhance database operations with AI and allow users in implementing AI solutions with Db2. Shaikh also manages research collaborations between IBM and Canadian universities. He and his team have created several Python frameworks to optimize complex data science tasks like exploring high dimensional datasets and improving data quality. Additionally, Shaikh has 12 scientific publications, 5 issued patents, and 11 filed patents. He completed his master’s degree at the University of Waterloo and is currently pursuing a Ph.D. in machine learning at York University.