A few years back I published a piece for TechTarget that stacked App Attach up against the rest of the app-layering crowd, and how the “old” MSIX AppAttach service worked. Now Microsoft has made several changes both in their application virtualization stack but also the AppAttach service has gotten some improvements. Microsoft has reversed course …
After working a lot with GenAI the last couple of years, I see that some organisations must keep data, LLM inference inside their own data centres for compliance, latency or cost reasons. Therefore I wanted to show how to build a PoC/MVP running locally but also look more into the enterprise ecosystem in terms of …
Running Large Language Models on Private Infrastructure Read More »
Last week I noticed that Merill Fernando published a MCP Server for Entra ID / Graph called Lokka. I wrote earlier this week about MCP https://msandbu.org/making-actions-available-instead-of-apps-with-mcp/ and how it allows actions and APIs to be easier available using natural language. This allows to get a bunch of commands available from a MCP Host such as …
Setting up MCP Server to Entra ID / Graph with Lokka Read More »
For those that have worked with Citrix, MicroApps might ring a bell? The product they had allowed you to create Microapps, which was these small components that was running API calls to the backend-service and making it easy to present these on a front-end. While that product didnt work out great for Citrix, I loved …
Making Actions available instead of Apps with MCP Read More »
Since the initial release of Copilot for Microsoft 365 I have been bombarded with questions related to many different areas about the service. Therefore I decided to create a summary about many of the questions that I get. Some of them are basic, some of them are deep-dive. Hopefully some of you will find it …
Already two years ago I wrote a blog post on how different LLMs are benchmarked, safe to say that much has happened in the last two years in terms of how these models no are benchmarked. Since many of them have also gone past the initial scoring of many of them and we have needed …
Five days ago, the Chinese company DeepSeek launched its new GenAI model, DeepSeek-R1. This model has quickly taken the GenAI world by storm, as it has been shown to compete with OpenAI’s latest models, including the more “thinking models” OpenAI-o1, at a fraction of the price. While it “only” has 671 billion parameters, rumors suggest …
This blog post is essentially a long recap from my session at the Nordic Infrastructure Conference in November (which you can find the presentation here https://github.com/msandbu/nic2024) In this session I talked about * The core networking stack and some numbers in Azure * Features and the software-defined networking stack and how to troubleshoot* How many services in Azure …
Whilst working with GenAI the last years, much of the focus the last year has been on GenAI Agents. So what is a GenAI Agent? Well there are a couple of characteristics that define it, while the line between a virtual assistant and a GenAI agent might be small there are some minor details to …
This is a blogpost that I have been wanting to write for a while, and since I have gotten asked a lot of questions throughout the years on this topic. How do I stay up to date on relevant topics? how do I learn new things? Therefore I wanted to write down my approach, does …
Mastering new skills and staying ahead in Technology Read More »
