Oh yes, jumping on the band wagon again!

Open AI released a whole lot of updated models and so on last week. But one tool (in preview) jumped out. Open AI Codex CLI.

Now, from first impressions, this looks and feels a lot like Anthropic Claude Code that I looked at a few weeks back. But hey, everyone copies everyone right?

Anyway, it is more of a “technical” thing to use but you really don’t have to be super technical. It, like Claude Code, generated very clean code based on just a couple of prompts.

Anyway, take a listen for more :)

Some links from the show:

Open AI Codex CLI - Getting Started

Hey, subscription to this podcast is FREE if you want to subscribe! I hadn’t realised before it forced you to pay ;) Of course, if you wish to or must pay, then please feel free with a paid subscription or via Patreon

Or just buy a unicorn a coffee here!

Oh, and yes, I have ended up on YouTube (doesn’t everyone eventually?):

https://www.youtube.com/@justfifteenmins but don’t worry, no ugly face to worry about (yet!).

Unicorn University:

https://applied-ai-services.kit.com/020c6040f2

I have already burdened you with an episode about MCP. Now, I am digging into some actual code examples in dotnet (maybe I will try other languages later). What the heck is an MCP host? MCP client? MCP server?Well, take a listen to find out more!

A few links mentioned in this episode:

Microsoft partners with Anthropic to create MCP SDK

GitHub Repo of the MCP SDK and examples

Hey, subscription to this podcast is FREE if you want to subscribe! I hadn’t realised before it forced you to pay ;) Of course, if you wish to or must pay, then please feel free with a paid subscription or via Patreon

Or just buy a unicorn a coffee here!

Oh, and yes, I have ended up on YouTube (doesn’t everyone eventually?):

https://www.youtube.com/@justfifteenmins but don’t worry, no ugly face to worry about (yet!).

Unicorn University:

https://applied-ai-services.kit.com/020c6040f2

Indeed!

Azure Functions has been around for some time. Like AWS Lambda, it allows you to create the smallest piece of functionality for whatever code you need to build i.e. an API, a timer-based execution function and so on. Although you can build an extensive system with Azure functions, you can quickly become embroiled in some spaghetti code and communication between functions (similar to microservices etc.). Now, there is something called Azure Durable Functions - which I have to confess it has been a while since I looked into this, but the key aspect of durable functions was to orchestrate flow and so on between Azure functions.

Which now begs the question, using .NET Aspire within Azure Functions supersedes durable functions?

I am not sure, as I haven’t delved too deeply, but it certainly looks like a more modern approach to build distributed systems and, in this case, with Azure functions. This is not only available when building new Azure Functions, you can add the Aspire orchestration to existing functions too.

Hey, subscription to this podcast is FREE if you want to subscribe! I hadn’t realised before it forced you to pay ;) Of course, if you wish to or must pay, then please feel free with a paid subscription or via Patreon

Or just buy a unicorn a coffee here!

Oh, and yes, I have ended up on YouTube (doesn’t everyone eventually?):

https://www.youtube.com/@justfifteenmins but don’t worry, no ugly face to worry about (yet!).

Unicorn University:

https://applied-ai-services.kit.com/020c6040f2

Well, love it or hate it, consumer-based AI app building is here. I haven’t had chance to try Loveable as yet, but for today, we have Replit.

Replit is a new kind of online app building environments that you can use in your browser (or even from a mobile phone) without having to download any editor/development environment. Replit also takes the app building further by hosting your app and even allowing you to make money (bounties!) by selling your app or access or selling your services to help others build on their creations.

Certainly, as an all-in-one browser environment, it is pretty impressive, and I need to give myself more time to properly explore!

It is a strange new world building apps with AI but i think augmented development is here to stay for better or worse!

Hey, subscription to this podcast is FREE if you want to subscribe! I hadn’t realised before it forced you to pay ;) Of course, if you wish to or must pay, then please feel free with a paid subscription or via Patreon

Or just buy a unicorn a coffee here!

Oh, and yes, I have ended up on YouTube (doesn’t everyone eventually?):

https://www.youtube.com/@justfifteenmins but don’t worry, no ugly face to worry about (yet!).

Unicorn University:

https://applied-ai-services.kit.com/020c6040f2

Who would have thought? Tron Unicorns! oh yes!

The Model Context Protocol (MCP) is an open standard designed to connect AI models with external tools, data sources, and services. Think of it as a "USB-C for AI applications"—a universal interface that simplifies how AI systems interact with various resources. MCP enables seamless integration, making it easier to build workflows and enhance AI capabilities.

Quick Overview of MCP:

Purpose: MCP standardizes how applications provide context to AI models, enabling them to fetch data, call tools, and interact with services.

Architecture: It follows a client-server model:

MCP Hosts: Applications like IDEs or AI tools that need access to data.

MCP Clients: Protocol clients that connect to servers.

MCP Servers: Lightweight programs exposing specific capabilities (e.g., accessing databases, APIs, or local files)

Flexibility: MCP supports both local and remote data sources, making it versatile for various use cases.

A few links for MCP “servers” (remember, these are not OS servers, but lightweight “server” programs using MCP to connect to local/remote things like File Systems, YouTube or whatever):

Awesome MCP Servers

VS Code MCP Extension

MCP Server Finder | The Ultimate Model Context Protocol Directory

MCPServers.Net | Find Your Ideal MCP Servers & Resources

Hey, subscription to this podcast is FREE if you want to subscribe! I hadn’t realised before it forced you to pay ;) Of course, if you wish to or must pay, then please feel free with a paid subscription or via Patreon

Or just buy a unicorn a coffee here!

Oh, and yes, I have ended up on YouTube (doesn’t everyone eventually?):

https://www.youtube.com/@justfifteenmins but don’t worry, no ugly face to worry about (yet!).

Unicorn University:

https://applied-ai-services.kit.com/020c6040f2

What to say?

Well, thanks to Tim Abell for all of the JFM suggestions - I will get to them I am promise! You can also take a listen to software should be free (and actually, part of this ramble is about just how much free resources there is!).

Hey, subscription to this podcast is FREE if you want to subscribe! I hadn’t realised before it forced you to pay ;) Of course, if you wish to or must pay, then please feel free with a paid subscription or via Patreon

Or just buy a unicorn a coffee here!

Oh, and yes, I have ended up on YouTube (doesn’t everyone eventually?):

https://www.youtube.com/@justfifteenmins but don’t worry, no ugly face to worry about (yet!).

Unicorn University:

https://applied-ai-services.kit.com/020c6040f2

I hadn’t realised “Play misty for me” was a crazy phycological thriller (apparently, the first film Clint Eastwood directed). But I just remembered the song actually!

Anyway, similar to Ollama, LM Studio, Msty looks like a very easy option to run LLMs locally (as well as accessing larger models up in the cloud like Open AI etc.).

Hey, subscription to this podcast is FREE if you want to subscribe! I hadn’t realised before it forced you to pay ;) Of course, if you wish to or must pay, then please feel free with a paid subscription or via Patreon

Or just buy a unicorn a coffee here!

Oh, and yes, I have ended up on YouTube (doesn’t everyone eventually?):

https://www.youtube.com/@justfifteenmins but don’t worry, no ugly face to worry about (yet!).

Unicorn University:

https://applied-ai-services.kit.com/020c6040f2

Well, another day another AI tool - for coding anyway.

I had a go at building a cross-platform mobile app using GitHub Copilot. It did eventually build something I wanted and better still, using best practices. But it did take some patience!

I have the fortunate (or unfortunate) position to not know when to give up ;)

So, also tried Claude Code. Bit more of a developer tool being a command line tool, but still pretty easy to get going. You do need an active Anthropic account (the people who created Claude Sonnet/Code) and at least $5 of credit. In the app I created, this was less than 5c! Try outsourcing that for that much!

Anyway, more of a weekend ramble this one! rather odd - I have a YouTube channel called “Just Fifteen Mins” but I think I might have to swap around the video and podcast channel at this rate!

Hey, subscription to this podcast is FREE if you want to subscribe! I hadn’t realised before it forced you to pay ;) Of course, if you wish to or must pay, then please feel free with a paid subscription or via Patreon

Or just buy a unicorn a coffee here!

Oh, and yes, I have ended up on YouTube (doesn’t everyone eventually?):

https://www.youtube.com/@justfifteenmins but don’t worry, no ugly face to worry about (yet!).

Unicorn University:

https://applied-ai-services.kit.com/020c6040f2