Issue 21: Doing That Rag!

🗞️ AI News of The Week:

Google Drops "Nano Banana" Model 🍌

Google took the AI-Nerd World by storm with the release of Nano Banana after teasing it out on LM Arena, where AI-detectives traced the source of the model to Google.

Nano Banana is a next-generation AI image editor developed by Google, designed to transform and edit images using simple natural language prompts with a focus on character consistency and high-quality scene blending.

It’s standing out for how well it can modify or change parts of images you upload while keeping everything else in the image the same (swapping out people, objects, etc).

👉 Try it out here or on Google Studio

🧠 AI Term of The Week: Text Chunking

Text Chunking is the process of splitting large documents or transcripts into smaller, manageable pieces, or "chunks" so that RAG systems (Retrieval-Augmented Generation) can process them easier.

For example:
I just built my first RAG System (basically a personal knowledge base powered by AI) and I fed my RAG a big text document that had 145,000 characters in total. I ran a simple Python Chunking Script and it ‘chunked up’ my document into 350 chunks so the AI could work with the document more efficiently.

💬 Quote of the Week:

🧰 AI Tool of The Week: WisprFlow (aka, Flow)

WisprFlow or Flow, is an amazing AI-powered voice dictation tool that lets you just talk (or whisper) into your microphone or phone and text appears magically and the accuracy is crazy.

You just enter a keyboard shortcut to activate Flow and start talking instead of having to type on a keyboard. Flow works in pretty much every application and syncs seamlessly across devices. Available on Mac, Windows, iPhone.

👉 Check Out Flow

🛠 How I Built a RAG System to Mine My AI Chat History for Gold

I built my very first RAG! It’s nerdy, but pretty awesome. RAG is the ‘AI Thing’ that I will be exploring in great detail as I think it’s what most creators and small businesses want (using AI with their own documents) and there are a million ways to build them and scale them and I’m into them. RAG makes it possible to give AI access to all of your documents so you can find things, create things, brainstorm, etc and you can keep it totally private and secure.

In this tutorial, you will learn about RAG (Retrieval Augmented Generation) and how to build a RAG system from scratch that mines your AI Chat Conversations for ‘gold nuggets’ of information.

What You Will Learn:

• What is RAG (Retrieval Augmented Generation)?

• How Does RAG Work?

• How To Build A RAG System Using ChatGPT Conversations?

• What Are The Prerequisites?

• How To Use VS Code & Terminal To Run Python Scripts

📺 Watch Tutorial on YouTube

Mike Murphy — The AI Handyman 🧰
Helping creators & small businesses turn their content & documents into AI‑powered tools.

• 📚 Nano Banana

• 🛜 Text Chunking (RAG)

• 🧰 Flow

• ▶️ How I Built A RAG System To Mine for AI Chat History Gold

• 💬 “The only limit to our realization of tomorrow

  is our doubts of today.”– FDR