Microsoft Reactor Stockholm

Join us for the first session in our Python + AI series!

In this session, we'll talk about Large Language Models (LLMs), the models that power ChatGPT and GitHub Copilot. We'll use Python to interact with LLMs using popular packages like the OpenAI SDK and Langchain. We'll experiment with prompt engineering and few-shot examples to improve our outputs.

We'll also show how to build a full stack app powered by LLMs, and explain the importance of concurrency and streaming for user-facing AI apps.

In our second session of the Python + AI series, we'll dive into a different kind of model: the vector embedding model.

A vector embedding is a way to encode a text or image as an array of floating point numbers. Vector embeddings make it possible to perform similarity search on many kinds of content.

In this session, we'll explore different vector embedding models, like the OpenAI text-embedding-3 series, with both visualizations and Python code. We'll compare distance metrics, use quantization to reduce vector size, and try out multimodal embedding models.

AI in Microsoft Fabric is designed to help you work smarter with your data. In this session, you will learn how Copilot accelerates analytics by generating queries, building models, and creating visualizations using natural language.

We will explore AI functions in notebooks, which make it easy to apply enrichment such as summarization, sentiment analysis, and entity extraction with a single line of code. You will also see how AI-powered transforms on OneLake shortcuts can automatically process text files for summarization, translation, or PII detection without building pipelines.

Finally, we will show how Data Wrangler provides an interactive way to clean and shape your data inside notebooks while generating reusable code.

Through a live demo, you will see how these capabilities combine to help you deliver insights and intelligent features quickly for your hackathon project.

What you will learn

• How to use Copilot for modelling and analytics
• How to enrich data with AI functions and shortcut transforms
• How to prepare and transform data efficiently with Data Wrangler

In our fourth Python + AI session, we'll explore one of the most popular techniques used with LLMs: Retrieval Augmented Generation.

RAG is an approach that sends context to the LLM so that it can provide well-grounded answers for a particular domain.

The RAG approach can be used with many kinds of data sources like CSVs, webpages, documents, databases.
In this session, we'll walk through RAG flows in Python, starting with a simple flow and culminating in a full-stack RAG application based on Azure AI Search.