AI Setup (2026)
This document is and will probably remain, incomplete. I am using it to document some local AI setup items for 2026, as a reminder so I can see how this shifts over time.
[Read More]
Being a Technical IT Leader
When leading a large technical team, I’ve found that all things are circular. As you slowly tweak for people and problems, you often find yourself implementing “fixes” that you had before, but slowly diverged away from. This list serves as a reminder of all the things that actually work -...
[Read More]
Running coding models on an Apple Mac mini M4
With the price of tokens set to soar, and coding with AI now the norm, running local models is becoming increasingly important to minimise costs.
[Read More]
Setting up replication for MongoDb v4
This post is a reminder to myself for setting up replication for MongoDb for Disaster Recovery, as all the AI agents seemed to have this wrong and/or missed a number of steps. This should work for versions 5 and 6 of MongoDb, but for the purposes of this article it...
[Read More]
Setting up a new machine with winget
I got a new laptop for work, and decided to take the future pain out of setup using WinGet. WinGet comes
baked into Windows version 24H2 and above, and is a convenient way to keep all your applications up to date.
[Read More]
Vertical Slice Architecture
This post is as a result of watching a video on YouTube presented by Chris Sainty at NDC Oslo.
[Read More]
Retrieval-Augmented Generation (RAG) using Functions with Mistral and .NET
This blog is the third of three, where I decided to try out Mistral.AI. This article covers how to use Mistral’s large language models (LLMs) in conjunction with functions that are exposed to Mistral. The LLM will then choosse to execute/use those functions if it feels they provide additional context...
[Read More]
Retrieval-Augmented Generation (RAG) with Mistral and .NET
This post is the second of three, where I decided to try out Mistral.AI. This article covers how to use Mistral’s large language models (LLMs) to use Retrieval-Augmented Generation (RAG) to query your own documents.
[Read More]
Storing embeddings in PostgreSQL using Mistral and .NET
This post is the first of three, where I decided to try out Mistral.AI. This article covers how to use Mistral’s large language models (LLMs) to create embeddings from your own documents using .NET.
[Read More]
Adding user name and email address to an Auth0 access token
When creating a Single Page Application (SPA) with a backend API using [Auth0], there are two types of tokens you will encounter:
[Read More]