Running inference against LLM models on your own local hardware means no API costs, no data leaving your machine, and no vendor lock-in. In this workshop, Bill will introduce Kronk, a Go SDK that lets you embed local model inference directly into your Go applications with full GPU acceleration — no CGO required.

Whether you are building chat, vision, audio, embedding, or tool calling applications, Kronk gives you the same power as a model server without needing one. You will see live demos from writing your first chat app to driving a coding agent with a local model.

With that solid foundation, you’ll build real applications:

• Retrieval-Augmented Generation (RAG) pipelines that ground model responses in your own documents using embeddings and vector search.
• Natural language-to-SQL systems where the model generates database queries from plain English, with grammar constraints ensuring the output is always valid executable SQL.

By the end of the day, you’ll have built applications that load models, cache intelligently, retrieve context, and generate code — all running locally on your own hardware.