Rob Kerr's Blog

NVIDIA DGX Spark

These videos cover configuration, use and thoughts around the NVIDIA DGX Spark platform

5 videos

Building a GraphRAG App: Concepts, Code, and Demo 18:10

Building a GraphRAG App: Concepts, Code, and Demo

NVIDIA DGX Spark

In this video, we'll explore what GraphRAG is all about and why it matters for building smarter AI apps. We'll walk through the core concepts, from understanding graph databases and how they model relationships, to seeing exactly how GraphRAG improves LLM retrieval with structure and context. Then, we'll dive into a full code walkthrough of a real application, showing how the backend, database, and UI all work together to make your AI more accurate and explainable.

Nov 27, 20252,827 views
Vectorizing Images with LLMs [Full Stack Deep Dive] 13:04

Vectorizing Images with LLMs [Full Stack Deep Dive]

NVIDIA DGX Spark

In this video, we'll walk through how to build a complete end-to-end image vectorization and similarity search system using modern AI tools. We'll explore how to generate image embeddings with a multimodal LLM, store them efficiently in ChromaDB, and see how LLM microservices power the semantic search logic—all while checking out the architecture and watching a live demo in the browser.

Nov 20, 20251,461 views
From Text to SQL: Fine-Tuning Phi-4-Mini-Instruct with LoRA and PyTorch (Deep Dive) 37:19

From Text to SQL: Fine-Tuning Phi-4-Mini-Instruct with LoRA and PyTorch (Deep Dive)

NVIDIA DGX Spark

In this video, we'll walk through the full workflow of fine-tuning the Phi-4-Mini-Instruct model using LoRA and PyTorch to boost natural-language-to-SQL accuracy. We'll explore how fine-tuning compares to RAG, then dive into coding the training script, setting up Docker, and quantizing the model so it runs efficiently on local devices. By the end, we'll have a customized small language model ready to test on a desktop.

Nov 09, 20251,027 views
Running OpenAI’s GPT-OSS-20B Locally with Open WebUI (Full Setup Guide) 5:10

Running OpenAI’s GPT-OSS-20B Locally with Open WebUI (Full Setup Guide)

NVIDIA DGX Spark

In this video, we'll walk through how to set up Open WebUI with OpenAI's GPT-OSS-20B running locally inside Docker. We'll show you how to create a container with Llama.cpp, expose Open WebUI to your system via an NVIDIA Sync SSH tunnel, and download and run the 20-billion-parameter model on your own hardware. Whether you're using a desktop-class GPU or following along on a Windows or macOS laptop, we'll guide you through everything you need to get started with local AI development.

Nov 03, 20256,343 views