Coffee & Code #2: one of my favorite publicly available healthcare datasets is MEPS. It doesn't have risk scores however, so I bolted those onto it. Allows me to do some interesting analyses.
For example, looking at the growth of GLP-1 especially as it relates to underlying morbidity as measured by concurrent risk scores. Use and morbidity trend very differently. Nothing too surprising in here, still good to line up expectations against data.
The "code" part? Trying to improve the infographic capabilities of my custom harness. Visualization using matplotlib, even with expensive models is still hit or miss currently (though I'm using a very cheap one here). If you have found something that has worked well for you, share!
The "coffee": 2-shot latte, half oat, half whole.