About the event
AWS and Weaviate, a leading Vector Database platform, bring you a hands-on workshop where you’ll learn about Amazon Bedrock and Weaviate. Learn how Weaviate, an open source vector database, reduces the complexity of building and scaling AI applications.

Explore the Weaviate platform and learn how to build AI-native applications from the ground up. Level up your skills with the latest vector search and RAG techniques. You will hear what is new and best practices from the team that built it.

Who should attend
Ideal for developers looking to enhance their expertise in AI-Native applications, offering a blend of theory and practical exercises. Whether you're starting from scratch or aiming to polish your skills, this workshop will equip you with the knowledge and hands-on experience to innovate and excel in the field.

Requirements:
Bring your laptop. You will need:

• Comfortable with Python and Jupyter Notebooks.
• Basic understanding of (NoSQL) databases.
• Dev Environment – you will use a cloud-based IDE, so a local setup is not required.
• AWS knowledge – nice to have

Full agenda
Intro – Introduction to AI-Native Applications
13:00 - 15:30 CEST

Dive into the core concepts and the building blocks of AI-Native applications, including ML models, vectorizers, and vector databases. Discover their capabilities and why they are so fast.

• Lesson 1 – Initial Setup & Hello World Learn how to create a new Weaviate Instance, set up your AWS account to work with Amazon Bedrock models, use Bedrock models with Weaviate, and make the first call from Python.
• Lesson 2 – From 0 to your first ML-based search You will work through an end-to-end process – to learn how to load data with a Bedrock Model (using a vectorizer), run various queries, and apply filters to refine your search results.
• Lesson 3 – Bring Your Own Vectors In some cases, you may already have vector embeddings generated. Learn how to load the data with vector embeddings, run pure vector queries, and use vectorizers on your vectors.

Enhancing Search with Generative AI (RAG)
15:30 - 17:30 CEST

• Lesson 4 – Enhancing Search with Generative AI (RAG)
  Explore how to augment your search results by generating answers from user prompts using popular Bedrock Generative models. Gain insights into integrating generative AI to better understand your data.
• Lesson 5 – Implementing Multi Tenant applications Learn how to separate your data into isolated buckets (tenants), while keeping the same structure across all of it. This is handy when you work with data from multiple users, but you need to keep it separate, or maybe you work with many PDFs, and you want to search on each of them independently. The possibilities are endless. You will learn about use cases and how to implement one of them.
• Lesson 6 – Multivector configuration By default, vector databases use one vector embedding per object. However, you could look at your data from different angles and generate vectors for different properties or even mix different Bedrock models. Learn how, why, and when to multiple vectors and how to query separate vector spaces.
• Bonus Lesson – Vector Compression Learn about vector compression algorithms and strategies that affect recall, performance, and cost-efficiency of vector databases.

Networking and Drinks
17:30 - 18:30 CEST