Databricks Step by Step: Build a RAG Chatbot

A step-by-step guide to building a retrieval-augmented generation chatbot on Databricks with chunked documents in Delta, an AI Search index, and a hosted foundation model.

Databricks Step by Step: Serve a Model as an API

A step-by-step guide to deploying a Unity Catalog model to a Databricks Model Serving endpoint, calling it over REST and Python, updating versions safely, and monitoring it.

Databricks Step by Step: Train and Register a Model

A step-by-step guide to training a machine learning model on Databricks, tracking experiments with MLflow, and registering versioned models in Unity Catalog with aliases.

Databricks Step by Step: Build a Data Pipeline

A step-by-step guide to building a bronze, silver, and gold data pipeline on Databricks with Auto Loader, Lakeflow pipelines, data quality expectations, and a scheduled job.

The Hand

A short story by Henry Bernreuter.

The Vibe-Coding Trap: What Building an AI Project Manager Taught Me the Hard Way

A first-person account of building an AI project manager that writes and reviews its own code, and the recurring gap between "the AI says it's fixed" and "it's actually fixed" that nearly sank it.

Vector Databases in One Post

Vector databases exist because traditional databases can't answer the question 'what is similar to this?' Here's the problem they solved, how they work, and when to use them.

AWS Bedrock, Azure AI Foundry, and Vertex AI in One Post

AWS Bedrock, Azure AI Foundry, and Google Vertex AI are the three cloud platforms fighting for the enterprise AI market. Here's what they are, how they got here, and how to choose.

MCP and A2A in One Post

Model Context Protocol and Agent-to-Agent are the two protocols trying to solve AI interoperability — one for tools, one for agents. Here's what they are and why they exist.

Make, n8n, and Zapier in One Post

Zapier, Make, and n8n each solved the problem of connecting apps without code. Now all three are racing to become the backbone of AI agent infrastructure.

LangGraph in One Post

LangGraph is the stateful, graph-based framework that LangChain built when it realized real agent workflows need loops, not lines.

Run Your Own Model or Use the API? A Practical Guide

The honest cost breakdown for self-hosting LLMs vs. using Claude or another API — privacy, latency, capability, and the math that actually drives the decision.

Agent vs. No-Agent: A Production Decision Guide

The real cost of building agents in production — failure modes, debugging nightmares, and the five questions that tell you whether you actually need one.

How to Architect an AI ETL Pipeline

A practical guide to designing AI-powered ETL pipelines — where LLMs actually belong in extract, transform, and load, and where they just add cost.

When to Build an Agent (And When You're Overcomplicating It)

A practical decision framework for choosing between raw API calls, orchestration frameworks like LangChain, and full agentic systems — with real examples from production.

Databricks in One Post

Databricks was founded by the team that created Apache Spark. The lakehouse architecture they invented is now how most enterprise data and AI work gets done.

TensorFlow in One Post

Google Brain needed a framework to run AI at search-engine scale. What they open-sourced in 2015 became the reason deep learning left research labs and entered every industry.

FastAPI in One Post

Sebastián Ramírez was a freelance developer in Colombia who needed faster, cleaner API code. What he built in 2018 became the backbone of AI inference infrastructure worldwide.

Hugging Face in One Post

Three French founders pivoted from a Gen-Z chatbot to hosting AI models. The Transformers library they built is now where every major model in the world lives.

LangChain in One Post

Harrison Chase released LangChain six weeks after ChatGPT. The core insight — that LLM calls need to be composable — reshaped how everyone builds AI applications.

Matplotlib in One Post

John Hunter was analyzing brain scans on a research budget. What he built to replace a $1,000 license became the bedrock of Python data visualization.

Pandas in One Post

Wes McKinney was frustrated by financial data. What he built in 2008 became the single most-used tool in data science.

NumPy in One Post

The origin story of NumPy — how one developer ended a civil war in the Python community and built the array library that became the language of AI.

Scikit-Learn in One Post

The origin story of the library that taught a generation to build machine learning models — and why it still matters in the age of ChatGPT.

Build a RAG Pipeline From Scratch in Python

A complete, copy-pasteable RAG pipeline using ChromaDB, sentence-transformers, and the Claude API. No LangChain. Every step visible.

What Is RAG and Why Does Every AI App Use It?

RAG is the technique behind every AI that actually knows your data. Here's what it is, why it exists, and why it matters — no PhD required.

I Accidentally Built My Own Publishing Workflow

How trying to simplify my blog turned into a full AI-powered publishing tool I actually use every day.

Linear Regression Example

Predicting Software Reselling Profits

Tweet War

Who will win the election according to Tweeter

Estimating COVID-19s transmission rates in Georgia

This is the most current transmisison rates for Georgia

Split Variables from one column into several columns in R

Cool Tool for R

Basic Data Visualiztion in Python

https://matplotlib.org/

Start Here

Install Python

Pandas in One Post

Intro into Pandas

Python in One Post

Basic python

How this site was built

How to build a site with Github and Jekyll

R in One Post

Basic R

Cool Data Visualizations in Python (code Included)

Numpy in One Post

Python Libraries