Serial entrepreneur, hands-on AI architect, open-source lover, always learning.

I build AI and platform systems that have to keep working once they meet production, compliance and scale. Over twenty years of it, so far.

Right now I lead AI platform architecture for the guest-facing AI systems at Four Seasons Hotels and Resorts. Before that I founded and ran a banking software company, led the build of a real-time tracking system for Canada's Department of National Defence, designed a travel booking platform, and started two IoT companies. I still like shipping software, not just architecting it, which is why FluentCMS and YeSvelte exist.

FluentCMS · Pre-1.0 · MIT license
An open-source CMS for .NET, built on ASP.NET Core and Blazor, that works as both a page-building CMS and a headless CMS on one API.
fluentcms.com · NuGet packages · Discord

YeSvelte · Pre-1.0 · MIT license
Svelte components whose look is a swappable theme: Tabler (Bootstrap) or daisyUI (Tailwind). Switch themes by swapping one stylesheet.
yesvelte.com

uBeac · Retired service · source published under the MIT license
An IoT platform I founded in 2017 that turned off-the-shelf Bluetooth gateways into live building dashboards in minutes. Acquired in 2020, and now published as a reference architecture.

Each project has its own page under Open Source on my site.

Smaller things, kept in the open:

• agent-framework-sample: a Blazor web application on .NET 10 with 24 interactive samples of the Microsoft Agent Framework, in six categories.
• claude-quickstarts-dotnet: .NET ports of two of Anthropic's Claude quickstarts, the customer-support agent and the autonomous-coding harness, rebuilt on the Microsoft Agent Framework.

Notes from my own research, for engineers building the same things: MCP servers, AI tooling for codebases, and the IoT and machine-learning work that came before.

• One MCP server, eight rulebooks: what the spec, Claude, ChatGPT, Codex, VS Code, Windsurf, Kiro and Gemini CLI require
  The checkable rules each AI host applies to an MCP server, from tool-name length to mandatory annotations and timeouts, where they conflict, and which tools validate against them. As of 1 October 2026.
• Testing MCP servers in 2026: 31 tools compared and the gaps between them
  Thirty-one tools that inspect, test, audit, secure or load test an MCP server, compared on twelve capabilities, and the gaps none of them fills. As documented on 1 October 2026.
• AI tools that document a codebase in 2026: 17 compared, and what none of them does yet
  Seventeen tools that turn a repository into documentation, from DeepWiki and Google Code Wiki to CodeWiki, OpenDeepWiki and Swimm, compared on access, outputs and audiences, with the gaps they leave. As of 1 October 2026.
• Observed, documented, inferred: a design for codebase documentation you can audit
  A design for documentation generated from code where every fact carries its evidence and one of three labels, observed, documented or inferred, so managers, business readers and coding agents can tell what was read from the code and what a model inferred.

The rest, back to 2017, is on pournasserian.com/writing, with an RSS feed.

pournasserian.com · LinkedIn · X