I have been experimenting with different ways to use AI, particularly beyond the familiar process of opening a chat, asking a question and copying the answer somewhere else. One project that caught my attention is Buzz, an open-source workspace designed to bring people and AI agents together. I wanted to explore whether I could use it to organise my work differently, with AI agents helping with research, writing and other tasks. Not just helping but actually doing the work. So instead of asking for an idea for a video shorts it will make the shorts for me. Or instead of asking for a title for a blog post I will wake up and it tells me the title, the research it did to know its a blog post worth adding and has written the post for me! Getting it running on Windows 11 taught me a few things about installing AI tools, connecting agents and managing background processes. ━━━━━━━━━━━━━━━━━━━━ WHAT IS BUZZ AI? Buzz is an open-source project from Block that provides a shared workspace for people and AI agents. Think of a collaborative messaging environment where you can organise conversations into channels and bring AI agents into those discussions. Rather than opening a separate AI conversation every time you need help, you can organise work around projects and give different agents specific responsibilities. For example, you might have: • A channel for content planning. • A channel for research. • A channel for software development. • Different AI agents assigned to particular tasks. The interesting possibility is having a more organised environment in which AI can contribute to your work, rather than treating every interaction as a separate conversation. ━━━━━━━━━━━━━━━━━━━━ WHY I WANTED TO TRY IT I have been using AI-assisted coding to build software, and I wanted to explore what happens when you move beyond asking AI to complete individual tasks. For example, I have content planning software. Rather than sitting down every morning and asking AI to generate content topics, I wanted to investigate whether I could develop a workflow where some of that work was prepared for me.

I have created documentation that outlines the steps you have to go through to install the self-hosted version of Djangify eCommerce builder on your own hosting. It looks like a lot, but actually it really isn't when you hand the documentation files I have included over to AI and let AI do everything for you. You'll spend most of your time just watching your AI scroll through everything and you won't really have to do much. You are responsible for anything that relates to payment. So adding your stripe information and your email settings set up. That's all classed as sensitive and secure information, so no AI will handle that. You have to add it yourself but other than that AI can do absolutely everything. That includes unzipping the folder after you've downloaded it and creating a new folder for you on your computer. Checking to make sure you have everything you need on your computer and walking you through anything you need to do step by step. As you go through each step you can tick it off. So it's all there and I hope it's useful https://www.djangify.com/docs/self-hosted-ecommerce-builder-install-guide/

Today I woke up and realised something. I built Djangify using a framework Django and Django uses HTML to build the front end - things like pages, templates and blog posts can take html code and display it. All an artifact or micro-app is, is html and CSS code blended together and AI is really good at making them look amazing. My idea was to make a blog post look and read better by giving everyone the ability to add tables, lists, checklist and just different things that will break a long article up, because long articles that help people are key for SEO. I wanted a way to present information without a long wall of words broken up by bullet points and titles. So I spoke to Claude code and came up with something and you can see what it looks like here https://www.djangify.com/blog/long-form-articles-seo-and-content-blocks/ and then inside of that article is a link to a live example. It shows the instructions to install Djangify self-hosted from beginning to end. It is responsive and looks great on a mobile and so from tomorrow I will be upgrading a lot more of my articles using different content boxes. Does it make a difference to my search traffic? That's the question I will answer through this experiment.

Just in case I haven't mentioned it the ability for guests to check out became available over the summer. Usually people had to register and validate their email first. Now they can add something to their cart, checkout and the system will create an account for them. Once their payment has gone through they receive an email to create their password and its all done. They are then shown the one time offer when they log in if you have one set up otherwise they are taken to downloads.

During the summer I added templates to the LinkHub and forgot to mention it. A LinkHub is a place to add links to what you want people to know about you then you use that one link on social media. You can include a video, your email sign up form or just link to your products or pages. Here's mine https://www.djangify.com/links/about-diane-corriette/ the images below show the different templates available. That's all in your admin area.