Building a Local-First AI Stack with Open WebUI & Ollama

AI is everywhere, but so are growing concerns around data privacy, vendor lock-in, and reliance on closed source models. Every prompt we send to cloud-based AI systems can potentially expose sensitive data, often with little transparency or control.

What if developers could run AI powered models locally, on their own machines, without sending data to external servers?

In this talk, I'll introduce a practical, local-first AI stack using Ollama and Open WebUI. We'll explore how these open-source tools enable developers to run, interact with, and customize AI models entierly on their own infrastructure.

Through a live demo, I'll show how to set up a fully functional local AI chat system, discuss real-world use cases, and highlight the trade-offs of running models locally, from prerformance constratints to hardware requirements.

By the end of this session, attendees will understand how to take back control of their AI workflows-building systems that are private, extensible, and truly their own.

Key Takeaways:

• Understanding the risks of relying on closed-source, cloud-based AI systems

• Learn what a local-first AI stack looks like in practice

• Get started with running LLMs locally using Ollama

• Build and interact with a self-hosted AI interface using Open WebUI

• Walk away with a practical setup to run your own private AI assistant