Topic 1 :
Title: Vector Databases Demystified: Powering AI with High-Dimensional Search
Description : Traditional databases are great at storing data. but what they store is raw information. AI workloads demand something richer - data that carries meaning, context and a sense of intelligence embedded into the data - called embeddings. These dense, high-dimensional data representations capture the essence of inferable content, storing and searching through embeddings requires a special purpose software called vector database. In this session, we will illustrate basics of vector databases, what makes them unique, when to use them, and why they’re essential for modern AI applications. We will look at some of the leading vector DB types like Milvus, Pinecone, and Weaviate. We will also dig into core indexing techniques like FAISS and HNSW, which enable fast, approximate nearest neighbour (ANN) queries at scale. Attendees will learn how to work with vector databases effectively and use it in most modern AI uses cases such as building RAG pipelines or LLM development.
Speaker Bio : HS Manoj Kumar is an intern at IBM with a passion for backend engineering and AI infrastructure. He enjoy’s working with technologies like Python, FastAPI, and vector databases to build scalable systems that power intelligent applications. His current focus is on exploring semantic search and retrieval-augmented generation (RAG) to make information access smarter and more efficient.
Topic 2 :
Title : Evolution of new Java features
Description : This talk covers the evolution of key Java features from Java 13 onwards, including pattern matching, virtual threads, sequenced collections, and flexible constructor bodies. We'll explore the evolution of these enhancements and how they improve code expressiveness, concurrency, and developer productivity, with live demos showcasing practical usage.
Speaker Bio :
Syed Moinudeen is a lead software engineer working with IBM Java Runtimes class library team at India Software Labs. He has 20+ years of experience in Java application development. Prior to joining IBM he has worked with Pervasive Engineering, American Online (AOL) organisations.
Vikas Chandra is a senior software engineer working with IBM Java Runtimes Class Libray team at India software Lab.He has 20+ years of experience in software design and development using C++ and Java. Prior to joining IBM, Vikas worked as a software engineer with various other organisations like Geometric Software, Unigraphics Solutions and Think3 Designs.
