This hands-on AI workshop demonstrates how to build a real-time chat application powered by large language models, streaming data, and retrieval augmented generation. Using Bytewax and Python, participants learn how to process live Slack messages, maintain conversation context, summarize discussions, and enrich responses with external data sources.

The session walks step by step through building a production-style pipeline that handles message routing, stateful conversation memory, windowing, and scalable real-time processing. You’ll also see how vector databases and embeddings are used to ground LLM responses in relevant documentation, enabling more accurate and context-aware chatbots.

You will learn how to:

• Build real-time LLM chat pipelines using streaming data
• Maintain conversation history and context at scale
• Combine live chat, vector search, and RAG for smarter AI assistants

As speakers we have Zander Matheson from Bytewax, and Henrik Nyman and Mikko Lehtimäki from Softlandia.

👉Watch the webinar below

‍

‍

Editorial note: The Bytewax company referenced in this post and webinar is no longer operating under its original name. Information related to Bytewax can still be found on GitHub.