Welcome
Hi, I’m Jamie. I’m an AI Engineer, Architect, Advisor and 8x Microsoft MVP. I’ve spent over 20 years in the software industry. I started this blog to document what I learn, both to reinforce my own understanding and to share it with the community, and it’s grown into a builder-in-public record of what I’m shipping, teaching and figuring out.
What I Do
For most of my career I’ve worked in and around the Microsoft ecosystem. Today my work falls into three areas:
- AI advisory: helping leadership teams shape AI strategy and product direction, decide what to build vs buy, and get from idea to production.
- AI engineering: building agentic AI systems with the Microsoft Agent Framework and Semantic Kernel, RAG pipelines, and MCP servers in .NET.
- Legacy modernisation: reverse engineering complex legacy systems and replacing costly components with custom .NET APIs to reduce running costs.
I split my time between client engineering work, advisory retainers, course authoring and writing.
Too Much Information
I’ve been writing code for years. It started when I’d get a magazine called Crash! fortnightly for the classic Spectrum. It had a tape with demos of full-price games (which I could never afford), so after I’d played those to death, I’d type out the code listings in the back of the magazine and play with those “games”. I’d hack away at them, changing how they behaved, and thought it was cool that I could control what they did.
This led to the Commodore 64 and a number of Amigas, where I hacked around with Blitz Basic and various incarnations of AMOS. During the Amiga years, I ran a BBS called The Forum! BBS from my bedroom, with people from all over the world dialling in over a 36k modem.
I ran a Fidonet node and was involved in the Public Domain scene. I collaborated with other coders in Europe and tried my hand at writing the next “Zelda”. We never finished it, but I’m sure I still have the sprites somewhere.
Naturally, all of this led to diplomas and degrees and a career in tech.
Machine Learning and Twitter
In 2013, I researched machine learning and sentiment analysis techniques during a Masters degree, which resulted in a product called Social Opinion that mines data from social networks. This work led to an invite to Twitter HQ in San Francisco in 2016.
In 2017, Twitter ran #Promote, which saw me extend the Social Opinion AI API to identify sales leads in real time. Digital marketers could then serve those leads with creative, delivering the right message, at the right time, to the right person. JiT (Just-in-Time) Ads was born!
This time I was invited to Twitter’s HQ in NYC. The competition for #Promote was fierce, and although I didn’t win (again!), the experience alone was worth it. I was granted access to internal alpha APIs with datasets that I integrated into my solution to give it further consumer insights.
That early machine learning work set the direction I’m still on: from sentiment analysis on social data to the agentic AI and RAG systems I build today.
Awards and Credentials
I documented my work and shared ideas on my blog and elsewhere, and in 2019 was awarded the Microsoft MVP Award in Artificial Intelligence. I’ve now held the award for eight consecutive years, currently in the AI and Foundry category. (Read more about the journey.)
I connected with 16 other AI MVPs from around the world and co-authored the first AI MVP Book, showing real-world applications of artificial intelligence. All proceeds went to charity.
I also hold the Microsoft Certified: Azure AI Apps and Agents Developer Associate (AI-103) certification, and I’m listed among Thinkers360’s Top 50 Global Thought Leaders in Agentic AI.
Teaching
I’ve authored 15 courses for Pluralsight, LinkedIn Learning and Cloud Academy, with a new Pluralsight course on integrating AI agents in enterprise systems on the way.
Current Focus
- Agentic AI systems, particularly with the Microsoft Agent Framework
- Production RAG pipelines and MCP servers in .NET
- Legacy modernisation and custom API development
- AI advisory and client engineering work
- New course content and demonstration-led writing on this blog
Work With Me
If you’d like help with AI strategy, an agentic AI or RAG build, or modernising a legacy system, I’d love to hear about it. You can book time with me on Calendly or connect on LinkedIn.
If you have questions or a topic you’d like me to cover on the blog, feel free to reach out.
~