-
Let's continue that .NET SDK (stream 381)
Summary
Fun stream: I walked through the latest Notebookmark release (v1.1.1), demoed the new customizable prompts and search filters, and showed how it helps prep my reading notes. Chatted about future tweaks (clickable links, vision SDK, better namespaces, tests) and shared thoughts on dev tooling, UI, and deployment options. Wrapped up energized for the next sessions.
šŗ - Twitch archive - stream no. 381
Replay
Read more...
-
AI and C# - It's time to wrap it-up and publish that version (stream 380)
Summary
Spent the stream working on NotebookMark, refactoring the search criteria class to integrate the AI-powered blog search prompts into the settings so users can customize themāyou know, the important stuff. We got it mostly working by the end, though apparently my laptop needed to stage a dramatic meltdown and reboot midstream to make that happen. At this point Iām pretty convinced my dev environment just enjoys watching me suffer, because clearly running Debian with .NET 9, Aspire, Docker (or Podman, whatever), and a dozen other things is just chefās kiss for stability.
šŗ - Twitch archive - stream no. 380
Replay
Read more...
-
Adding AI recommended blog to our "to read" stack (stream 379)
Summary
Wrapped up notebookmark work: upgraded Aspire, got AI-powered reading suggestions flowing, wired deletion/adding to the reading list, and surfaced settings for prompts plus favorite/blocked domains. Tweaked UI sizing, chased down missing publish dates, and wrestled with dev settings loading (because of course ādevelopmentā needed a capital D). And naturally, despite all that progress, my dev environment still found new ways to implode
šŗ - Twitch archive - stream no. 379
Replay
Read more...
-
Building an app to search across videos using AI (part 2) (stream 278)
Summary
Continued working on FrameFisher, successfully implementing video upload, delete, and search functionality using MudBlazor. Added a threshold slider to filter search results and integrated the API to search through video content using AI-powered vector search. Of course, I had to switch back to Debian from Omarchy because my dev environment kept breakingānothing says āprofessional streamerā like spending half your time fixing Linux distro issues instead of actually coding.
šŗ - Twitch archive - stream no.278
Replay
Read more...
-
Building an app to search across videos using AI (stream 277)
Summary
Started a new project called Frame Fisher to analyze videos using Reka Vision API, experimenting with MudBlazer instead of Fluent UI for a fresh UI look. Got the basic grid and service wired up, then tried migrating to the new .NET solution format and switching to .NET 10āat which point everything fell apart: VS Code couldnāt find the SDK, Rider had display issues, and reinstalling via different package managers left my environment in total chaos. Spent the rest of the stream rebooting, toggling between .NET 9 and 10, and wrestling with certificate trust issues on Arch, only to wrap up hungry and thoroughly annoyed that my āstableā dev setup decided to implode yet again. š
šŗ - Twitch archive - stream no.277
Replay
Read more...
subscribe via RSS