Training two models for better results in recommendation systems
Training one model for two different intents can lead to issues in recommendation systems. Two models trained on different signals can help to omit this issue.
Technical view of a recommendation system
A technical view on the RSS digest which provides insights into how a recommendation system works.
Selecting the features for the RSS digest
Beside of labels the inputs for an ML model need to be selected. This article gives insights into some thoughts and concepts.
Collecting labels to train a model for an RSS digest
This article shows how to collect and store labels for to train a model for an RSS digest.
Building an ML-powered RSS digest
Why I started an automatic digest for my feed subscriptions by using machine learning.
Temporary script, turned into a long-term solution
A case of 'there is nothing more permanent than a temporary solution'
I deleted my Windows installation
At least for now my Windows journey ends here after 22 years.
Agent Skills - a thin alternative to the Model Context Protocol?
The comparison between Model Context Protocol (MCP) and Agent Skills.
