Retro: Setting up a BBS Using Raspberry Pi Zero 2 and Mystic In the mid to late 90s, I ran a BBS on a Commodore Amiga 1200.  It started with 2 disk drives, and I eventually saved enough cash to get an IDE hard drive with a massive 520MB of storage. People from around the world would connect to that BBS on a single 36k modem.  About 80 registered users would dial into that BBS on a dedicated phoneline. I ran a Fidonet node that let user of the BBS send electronic messages globally.  This was before email and widespread internet access was available. Users would upload games, utilities, and public domain software.   There were no adverts. User data was not sold or analysed. There was no manufactured outrage. No algorithm to game. No Likes. Thumbs Up. Impressions or other vanity metrics. It ran on an isolated machine.     Screens were designed using ANSI graphics and menu navigation was simple. In this blog post, I document how to setup a BBS in 2024.  You will see how to use a budget Raspberry Pi, free software and services

[RETRO] Setting up a BBS Using Raspberry Pi Zero 2 and Mystic Learn how to use a Raspberry Pi, free software and services to host your own BBS at home for less than £51.

Read:
jamiemaguire.net/index.php/2024/08/17/ret...

[NEW COURSE]
Generative AI Stakeholder Alignment

My new course is live on Pluralsight!

Did you know 95% of organisations see little return on their AI efforts?

This course shows how to plan and manage AI projects that succeed.

Find it:
app.pluralsight.com/profile/author/jamie-mag...

First Look: Integrating Semantic Kernel with OpenAI What Is Semantic Kernel? Semantic Kernel had been on my radar for a while. Semantic Kernel is an open-source SDK designed to simplify the creation of agents that can interface with your existing code. The SDK is extensible and supports models from OpenAI, Azure OpenAI, Hugging Face, and more. Using the SDK means you don't have to learn the inner workings of LLM provider APIs. Semantic Kernel lets you build agents using C#, Python or Java.  Complex workflows and agentic AI orchestrations can be created using C#, Python or Java. With the open-source SDK from Microsoft recently reaching V1, now is the time to dig into what this can do. ~ Orchestration Layer Semantic Kernel sits at the centre of the agent stack.  It acts as the main orchestration layer for your agentic AI solution. AI models, and plugins (your existing code) can be easily integrated to help you create innovative solutions and user experiences. (image source Microsoft) In the past, this involved manually stitching together mul

[BLOGGED]
Integrating Semantic Kernel with OpenAI

Semantic Kernel is an open-source SDK designed to simplify the creation of agents that can interface with existing code.

Read:
jamiemaguire.net/index.php/2024/06/08/fir...

How I Used AI To Consume a 2 Hour Lex Friedman Podcast in 10 Minutes If you are reading this blog, you probably listen to one or many podcasts. Some of my personal favourites include Seth Godin’s Akimbo, Lex Friedmans, Mike Tysons, and Andrew Hubermans. Most of these podcasts span multiple hours, and despite wanting to listen to or watch them, I just don’t have that amount of free time. ~ The Idea This got me thinking.  I could use AI to create a tool that would use speech-to-text to transcribe the audio.  I could then use AI to perform document summarisation on the transcription.   This would let me skim read the summarised points to get a feel for the main points, and outcome of the discussion points in the podcast.   I could use the summarisations to help me decide if the podcast was worth listening to. ~ CLI Tool In this blog post, you will see how I built a CLI tool using a combination of Azure AI services to save me hours each week.   Specifically, you learn about: Creating a CLI tool that accepts a podcast URL Batch transcription

[BLOGGED]
How I Used Azure AI To Consume a 2 Hour Lex Friedman Podcast in 10 Minutes

Read more here:

jamiemaguire.net/index.php/2024/04/06/how...

ASP.NET Core | Open-source web framework for .NET Build web apps and services that run on Windows, Linux, and macOS using C#, HTML, CSS, and JavaScript. Get started for free on Windows, Linux, or macOS.

[SOURCE-CODE]
Bulk Alt-Text Generator

I hacked together a bulk alt-text generator web app,

A time saver if you need to bulk create alt-text for many images.

It uses ASP.NET and Azure AI services.

Code here:
github.com/jamiemaguiredotnet/AltTe...

20 Things I’ve Learned In 20 Years as a Developer. Whilst sipping my coffee and watching my 4 year old play his Switch I thought back to a time when I was a kid and enjoyed playing consoles. I looked back on the last few years and realised that this year is my 20th year as a professional in the trade. I started a thread on Twitter with things that were a common theme over the years and have exported this using Social Opinion and the conversation_id. 1. Learning never ends. This is both a curse and a blessing. I still love tech and get excited but I have days when I just want something to work without endless configuration or obscure configuration. 2. Everything is cyclical Do it on the client. Do it on the server. Do it on the client. Do it on the server.   Do whatever is best for the project, team and customers. 3. Everybody has opinions. That's cool. But sometimes the person with the loudest opinion doesn't mean it's the best. You have to listen to everything and everyone then weigh up the pros and cons of all suggestions or ideas. 4. Ask

I've been in the tech space for a while.

Here are 20 Things I’ve Learned In 20 Years as a Developer.


jamiemaguire.net/index.php/2021/10/16/20-...

My New Course With Cloud Academy Will Help You Pass the AI-102 (Implementing Conversational AI Solutions) Are you looking at sitting the new AI-102 Microsoft exam and gaining the Azure AI Engineer Associate accreditation? My new Cloud Academy course will help you understand the key concepts the exam covers in relation to the “Implementing Conversational AI Solutions” component of the exam. 15-20% of the exam content is covered by this topic. This new Cloud Academy course will teach you key components of building conversational AI applications and will cover the following learning objectives: Create a bot by using the Bot Framework SDK Using the Bot Framework SDK to create a bot from a template Implementing activity handlers and dialogs Using Turn Context Testing a bot using the Bot Framework Emulator Deploying a bot to Azure When Will The Course Be Released? This new course will be released next year. You can find out more details about what’s measured in other areas of the AI-102 here.

Looking to pass the AI-102 Microsoft exam?

My course helps you understand concepts in relation to “Implementing Conversational AI Solutions”.

Info here:

jamiemaguire.net/index.php/2021/11/07/my-...

Find a community you're interested in.

Serve it.

Listen to their problems.

Build something that helps them.

Audio Notes: Using Azure AI Speech to Perform Speech-to-Text Conversion In an earlier blog post, I introduced Audio Notes. This is a new a SaaS experiment that uses artificial intelligence, speech to text, and text analytics to automatically summarise audio and create notes from your recordings. In this blog post, I shared how you can use Azure AI Speech Services to perform continuous real-time speech-to-text transcribing. You will also learn about: speech to text transcribing options the JavaScript SDK main capabilities in the service key objects and events   A video demo is available along with full source code. ~ Azure Speech to Text Capabilities Real-time speech-to-text lets you transcribe speech from a microphone, file, memory stream. It can be used in scenarios such as: Creating, captions or subtitles in meetings Contact centres agent assistance Dictation ~ Single Capture Recognition Single-shot recognition lets you recognizes a single utterance. The end of a single utterance is determined by listening for silence at the end or until a maximum o

[BLOGGED]
Audio Notes: Using Azure AI Speech to Perform Speech-to-Text Conversion

See how to use Azure AI Speech Services to perform continuous transcription.


Read here:
jamiemaguire.net/index.php/2023/08/26/usi...

Semantic Kernel: Optimising Chat History When building agents using Semantic Kernel that integrate with SLMs (Small Language Models) /LLMs (Large Language Models), you typically send and receive multiple responses. A common pattern is to add all user and agent message to the ChatHistory object. In the past, I have done this by implementing a service class to handle these in .NET web applications. You can see an example of a controller which creates the chat history object for unique sessions for a webchat experience here: [HttpPost]   public async Task Post([FromBody] ChatRequest chatRequest)   {       // Use a fixed session ID for simplicity, or generate a unique one per user/session       var sessionId = default-session;       var history = _chatHistoryService.GetOrCreateHistory(sessionId);       // Add user input       history.AddUserMessage(chatRequest.Message);       var openAIProm

[BLOGGED]
Semantic Kernel: Optimising Chat History


Read here:

jamiemaguire.net/index.php/2024/11/23/sem...

How To: Using the Elasticsearch Vector Store Connector with Semantic Kernel I had been working with Elasticsearch to implement a site wide search feature and  an agentic AI system that will help developers find the right information at the right time for the task at hand. The content being searched and reasoned over is technical in nature. To support the agent AI feature, a vector database is required.  I did find an open-source Elastic connector project but could not get it to fully work without a lot of integration code. Fortunately, just a few weeks ago, Elastic released a Vector Store connector that is compatible with Semantic Kernel. In this blog post, I outline the steps you need to implement to use this new connector. The following is covered: Creating index mapping Creating an object with relevant attributes to represent vector content Generating content embeddings Chunking content Indexing content Representing the search query as an array of vectors Running the search using the array of vectors   I did have an issue with defining mappings

[BLOGGED]
How To: Using the Elasticsearch Vector Store Connector with Semantic Kernel


Read here:
jamiemaguire.net/index.php/2024/12/14/how...

From 4 Hours to 20 Minutes: How I Transformed and Consumed a Lex Fridman Podcast using Audio Notes AI Batch Transcription We’re all busy. Many podcast broadcasts regularly ship episodes that run for multiple hours.  Recently, a Lex Fridman episode was suggested to me on YouTube.  It was almost 4 hours long! It’s becoming increasingly common for podcasts to run for this long.  I don’t have that sort of free time to burn.  Chances are you don’t either. I decided to add a new feature to Audio Notes. Batch Transcription. ~ Audio Notes 1.0 If you’re not aware, the first version of Audio Notes let you create concise notes from the spoken word. You use the platform to record your voice, it then automatically transcribes your speech to text. The text is then summarised and saved. You can then use this for other tasks, or to perform a brain dump, or identify key topics. I dropped the ball as left v1.0 in the wild for a few months. Whilst on vacation I glance at the database and saw there were 1,960 sign ups!   Anyway, back to Batch Transcription. ~ Audio Notes 1.01 Saves You Time Batch Transcri

[BLOGGED]
From 4 Hours to 20 Minutes: How I Transformed and Consumed a Lex Fridman Podcast using Audio Notes AI Batch Transcription

Details:

Audio Notes: Optimising Batch Transcription In a previous blog post I introduced Audio Notes Batch Transcription. If you're not familiar with Audio Notes Batch Transcription, it helps save you hours of time each week, discover the key topics and sentences being discussed in YouTube content. This might be podcasts, interviews, technical sessions etc. It does this by using to AI automatically generate a transcript, associated abstract, and key sentences for any content that you are interested in. ~ Less Mouse Clicks. More Automation. The initial version of batch transcription involved several mouse clicks and manual processes to create the summarised video content.   In this recent update, I have automated most of these processes and reduced the number of mouse clicks. Back-end middleware has been created to automate processes in Microsoft Azure.  AI processes that generated NLP and text analytics insights are also automated. You no longer need to manually download the transcript and manually generate the summary. Those t

[BLOGGED]
Audio Notes: Optimising Batch Transcription


Read:

Hacking Around with GPT4o: Life Expectancy Visualiser It was my birthday recently.  It got me considering time. I hacked together a life expectancy visualiser from my cell phone. Built using GPT 4o in less than 5 minutes with a single prompt: Create an html page. It will let you supply your date of birth It will use national data to determine your life expectancy. When submit is clicked. It will render circles. 10 circles per row Each circle will represent 1 year. Years passed will be red. Optimal years will be green. Senior Years will be amber   Precise AI code output requires precise human input. I copied and pasted the code into an online tool on another tab, ran it and took the screenshot. Enter your date of birth and it renders a series of circles for each year of your life. Visualisation: Red: Years that have passed Green: Optimal active years Amber: Senior years And no, this is not my date of birth. Code free on GitHub here. Try it out using CodePen here. Time is the ultimate curr

[BLOGGED]
Hacking Around with GPT4o: Life Expectancy Visualiser

Details:

My New Course: Aligning Generative AI with Business Cases is LIVE! My new course “Aligning Generative AI with Business Cases” is LIVE on top training platform Pluralsight !   In this course, you’ll learn how to identify, develop, and evaluate suitable use cases for GenAI in your business:   First, you’ll explore how to evaluate and leverage GenAI for your business.   Next, you’ll discover the tools and concepts involved when developing GenAI solutions.   Finally, you’ll learn how to create an agentic AI solution that provides grounded, contextual, and relevant answers in natural language using the latest Microsoft and Open AI technologies.     When you’re finished with this course, you’ll have the skills and knowledge to help you identify, develop, and evaluate suitable use cases for GenAI in your business. ~ Demo Rich This no-fluff course is demo and code heavy.  It's aimed at an intermediate level audience. With 4 modules running just under 60 minutes, the course rapidly accelerates your learning. It shows how to sol

[BLOGGED]
Check my course “Aligning Generative AI with Business Cases” on top training platform @Pluralsight !


In this course, you’ll learn how to identify, develop, and evaluate suitable use cases for GenAI in your business.

Read more:

How To: Resolve Poor Performance on Raspberry Pi 4 When Running Visual Studio Code Occasionally, I will hack around on a Raspberry Pi 4 with my son.  Recently, we had been creating a fun home motion detection home security system. We had been using Thonny as it shipped with the Pi installation and was running smoothly. The simple interface was good for my son. I installed Visual Studio Code.  It was unusable. The performance was terrible. The slightest click or cursor movement took ages. The PI has 8GB of RAM so should be fine. After a little digging, it turns out, you need to disable hardware acceleration. ~ Visual Studio Code To do this, press F1 in VS Code, type runtime:   Select: Preferences, Configure Runtime Arguments The file will load, you need to set this to the following: That’s it. Visual Studio Code is now useable in the Pi. In a future post, I will share how we built an AI home security system using .NET, Semantic Kernel, Telegram, and Raspberry Pi. Stay tuned. ~  

[BLOGGED]
How To: Resolve Poor Performance on Raspberry Pi 4 When Running Visual Studio Code


Read more:

My New Course: Aligning Generative AI with Business Cases is LIVE! My new course “Aligning Generative AI with Business Cases” is LIVE on top training platform Pluralsight !   In this course, you’ll learn how to identify, develop, and evaluate suitable use cases for GenAI in your business:   First, you’ll explore how to evaluate and leverage GenAI for your business.   Next, you’ll discover the tools and concepts involved when developing GenAI solutions.   Finally, you’ll learn how to create an agentic AI solution that provides grounded, contextual, and relevant answers in natural language using the latest Microsoft and Open AI technologies.     When you’re finished with this course, you’ll have the skills and knowledge to help you identify, develop, and evaluate suitable use cases for GenAI in your business. ~ Demo Rich This no-fluff course is demo and code heavy.  It's aimed at an intermediate level audience. With 4 modules running just under 60 minutes, the course rapidly accelerates your learni

My course “Aligning Generative AI with Business Cases” is LIVE on top training platform Pluralsight !

In this course, you’ll learn how to identify, develop, and evaluate suitable use cases for GenAI in your business.

More details:

Courses – Jamie Maguire

[NEW COURSE]
Generative AI Stakeholder Alignment

95% of AI projects return little value - not a tech problem, a people and process one.

30 mins, 3 modules, and practical tools you can use on your next project.

Find it:
https://jamiemaguire.net/index.php/pluralsight-courses/

[OPEN SOURCE]
Visualising AI Job Exposure in the UK + EU

Karpathy's karpathy.ai/jobs maps Digital AI Exposure across US jobs. Excellent tool but US only.

I vibed a UK+EU version.

Blog:
jamiemaguire.net/index.php/2026/03/20/vis...

How I Use Audio Notes and the Microsoft Agent Framework to Save Hours Each Week – Jamie Maguire

[BLOGGED]
How I save hours each week using Audio Notes and the Microsoft Agent Framework.

Read:
jamiemaguire.net/index.php/2026/03/14/how...

I'm helping over 3.5K people each month upskill with AI on Pluralsight.

Whether you're a developer or decision-maker, we've got you covered.

Courses include: Gen AI Stakeholder Alignment, Vector Databases, and Crafting an AI Strategy.

Read:
app.pluralsight.com/profile/author/jamie-mag...

Audio Notes 2.0. Smarter Summaries, Better Insights, Less Time Wasted – Jamie Maguire

Feeling overwhelmed with endless content?

Every week, new podcasts, talks, and videos flood our feeds.

That’s why I built Audio Notes.

Just paste a YouTube or podcast link, and it gives you an AI-generated summaries, action items, and topic tags.

New Free Course: Understanding Microsoft Agent Framework – Jamie Maguire

[NEW FREE COURSE]
Understanding Microsoft Agent Framework

I have been shipping blogs, code, and videos as part of a FREE and self-published course on the Microsoft Agent Framework.

Details:
jamiemaguire.net/index.php/2026/02/24/new...

Microsoft Agent Framework-09:Adding RAG to Your AI Agent In this demo and blog, we take a different direction and look at how to give your AI agent access to your own documents using Retrieval Augmented Generation (RAG). Instead of relying solely on the LLM’s training data, the agent searches a vector store before each response and grounds its answers in your content. Why RAG Matters for AI Agents? LLMs are powerful but they have a fundamental limitation, they only know what they were trained on. They can’t access your internal documentation, product guides, company policies, or domain-specific knowledge. RAG solves this by retrieving relevant documents at runtime and injecting them into the model’s context. The result is an agent that answers questions using your data, not generic knowledge. This is useful when you want your agent to: - answer questions grounded in your data - cite sources so users can verify information - avoid hallucinations by constraining responses to known content RAG also helps your agent stay current without retraining the unde

[DEMO]
Microsoft Agent Framework: Adding RAG to Your AI Agent Using TextSearchProvider and In-Memory Vector Store

Watch:
https://www.youtube.com/watch?v=ykr-Y5IS0tI

Microsoft Agent Framework: From Personal AI Newsletter to Automated Email Broadcasts with AWeber – Jamie Maguire

[BLOGGED]
How to build an AI Email Marketing Agent with Microsoft Agent Framework 🤖

Full code walkthrough and demo:

jamiemaguire.net/index.php/2026/02/14/mic...

Microsoft Agent Framework-07: Exposing an Existing AI Agent as an MCP Tool In earlier posts in this series, we’ve explored how to extend AI agents using function tools, how to compose agents together as function tools, and how to build and debug MCP servers in .NET. In this demo, we look at how you can take an existing AI agent built with the Microsoft Agent Framework and expose it as an MCP tool, thereby making it available to any MCP-compatible client. This is where the two worlds collide. We already know how to build agents. We already know how to build MCP servers. Now we combine them. The idea is simple: take an agent, convert it to a function, wrap it as an MCP tool, and serve it over HTTP. Deep dive code and blog: https://jamiemaguire.net/index.php/2026/02/08/microsoft-agent-framework-exposing-an-existing-ai-agent-as-an-mcp-tool/

[DEMO]
Microsoft Agent Framework: Exposing an Existing AI Agent as an MCP Tool

This is where the two worlds collide.

We already know how to build agents.

We already know how to build MCP servers.

Now we combine them.

Watch the 5 minute demo:
https://www.youtube.com/watch?v=hoUuViiaIkc

Microsoft Agent Framework: Exposing an Existing AI Agent as an MCP Tool – Jamie Maguire

[BLOGGED]
Microsoft Agent Framework: Exposing an Existing AI Agent as an MCP Tool

This is where the two worlds collide.


Code and video here:
jamiemaguire.net/index.php/2026/02/08/mic...

Microsoft Agent Framework: Using Background Responses to Create an AI Researcher and Newsletter Publisher – Jamie Maguire

[BLOGGED]
Microsoft Agent Framework: Using Background Responses to Create an AI Researcher and Newsletter Publisher

Agent code and demo.

Read:
jamiemaguire.net/index.php/2026/01/24/mic...

Microsoft Agent Framework: Giving Agents Contextual Memory Using AIContextProvider In earlier blog posts we created a personal trainer agent.  It featured capabilities to process bookings search for and generate high protein recipes by using function tools and custom service classes. We saw how to implement human develop within the context of a payment scenario. In this blog post we look at how you can give agents access to memory across conversational threads. The following is covered: why give your agent memory overview of AI context provider how to persist agent memory how to fetch agent memory creating a function tool to calculate weight loss time serialising agent memory   We also look at events you can subscribe to when working with agent memory. We see how this can be used to give an LLM additional context or to ensure the LLM has all relevant information before processing a prompt. Code and a demo are included. Let’s dig in.   ~ Why Give Your Agent Memory When people interact with your agent it can be helpful to persist ce

[BLOGGED]
Microsoft Agent Framework: Giving AI Agents Contextual Memory Using AIContextProvider

Learn how to give agents access to memory.

Code and demo included:
jamiemaguire.net/index.php/2025/12/20/mic...