Back

Speaker "Sebastian Witalec" Details Back

 

Topic

Building AI-native apps with Vector Databases and Agents

Abstract

AI and Machine Learning are changing the world world, and we can observe the future with our own eyes. Every week, we get new ML models that promise to change our lives, and some of them are super impressive. But how do you bring them into your tech stack, your website/backend, to work with your data and provide AI-driven search and data augmentation, and even use agentic workflows to get the most of your data.
Who is this presentation for?
Techies who want to learn how to build AI-native applications that use Embedded models and LLMs to search and gain insight into their own data.
Prerequisite knowledge:
Comfortable with: * Python * Jupyter notebooks * (ideally) VS Code * basic GitHub
What you'll learn?
Join this workshop to learn core concepts like: what are LLMs, vector embedding, vector databases and RAG. This will take us 15 minutes, and we will dive straight into coding. We will build hands-on (so don't forget your laptops) an AI-powered app. We will cover: * Selecting an ML model * Loading and vectorizing your data * AI-based queries * Hybrid search (traditional keyword search combined with vector search) * Retrieval Augmented Generation * Usage of Agnets with Vector databases

Profile

Sebastian is a polyglot developer with a wide range of tech interests and a strong focus on tools that help us get the job done. His latest passion is in vector databases and AI-native solutions, helping developers build and productize AI-based applications to take advantage of the latest cutting-edge developments in Machine Learning. Warning: Talks involving Sebastian may contain pictures and interesting facts about meerkats and raccoons.