I've been developing my first desktop application "Athilio" which has been available for some time now.
Coupe of reasons why I built the app: 1. I wanted to do some user interfacing software outside of my "real job" where I basically only work with backend and infra 2. I wanted software where I could combine my data from my Garmin or Polar watch which I use to track my training sessions and my Oura ring which I use to track my daily activity and recovery 3. I wanted to try and build some features by applying LLMs (I know this reasoning is a bit backwards)
For the 2nd point there is already quite well established apps, but to me all of them honestly seem just crazy expensive. I would consider myself somewhat of a "lone wolf" fitness/sports person so I don't really want coaching or social features which many of these "training data aggregator" apps usually push on you. Also basically all of them are web/cloud services.
So my solution has no social features, all features work only around your own data, data is hosted on your own machine (sqlite). You could plan and view your training sessions with a Garmin watch completely offline with the app. User account is basically only needed for automatic downloading of your new data from your wearables service and a couple of AI features. For these reasons the paid tier is very affordable. Many of the concepts and usage flows are inspired by some common software observability solutions and concepts.
So some of the features are behind a paywall, free 14 day trial on sign up, no payment method needed.
The app is available for Windows: https://apps.microsoft.com/detail/9nqqc3qb5z52?hl=en-GB&gl=F... and for macos: https://www.athilio.com/downloads
Would be great to have feedback and feature suggestions!