Vector Databases in Action: Building a RAG Pipeline for Movie Recommendations - Part 1

AI robot aiming an arrow, symbolizing how a RAG pipeline targets relevant information for a user’s query. Using a movie recommendation agent to explore these architectural concepts Why I Am Building This Do you and your significant other ever sit down to watch a movie, and find yourself surfing previews for 30-45 minutes before giving up? Of course Netflix has a strong recommendation engine, but it’s tuned for maximizing engagement across their catalog, not necessarily reflecting your personal taste over time. ...

September 1, 2025 · 14 min · 2971 words · Mark Holton

Curiosity and Craft: What Drives Me to Build

Curiosity and Craft Exercising those muscles of grit and resiliency that are essential to improving at our craft I’ve been thinking about what drives me to build things. It is curiosity and craftsmanship. Those two things have been consistent throughout my career and life - from LabView as a Physics undergrad, to many different languages and technologies, to Rails web applications, to data pipelines to now LLM-based architectures. Even golf on the driving range as a kid, tinkering with clubs, swing changes. Trying to constantly improve and understand the fundamentals, and then put them to test. ...

August 16, 2025 · 7 min · 1292 words · Mark Holton

AI Integration: Building Conversational Apps with MCP

Why This Matters: The Shift to Agentic AI The software landscape is shifting toward agentic AI systems - applications where Large Language Models (LLMs) don’t just answer questions, but actively use tools to solve complex problems. Instead of building separate AI features, the market and industry are moving toward AI that can directly interact with your existing systems, databases, and workflows. ...

June 22, 2025 · 14 min · 2853 words · Mark Holton