Yeah getting it into something lightweight and runnable so the LLM can quickly iterate is key!

Hope you enjoyed and learned something!

I suspect no, they probably care more about the iceberg parts than the DuckDB parts? but this is purely speculative.

i've also been spending a lot of time on pg_duckdb (and how to run it) as we prepare for launching 1.0 of that extension. Can't wait to share more

!! Super clever

This practical example of how easy it is to bootstrap your observability is part of why I love working on DuckDB every day!

Another good strategy is partitioning your writes to object storage by date / time (i.e. by hour) so you can easily write naive queries for analysis and backfills, like `select * from 's3://my_bucket/year/month/day/hour/*.json'`

👀

i will never love another ERP like I loved Dynamics NAV.

There is nothing official I am aware of.

ps indeed this was initially designed to identify mispriced over/unders, although it has obviously expanded in scope since then

yes, fifa rankings applies an elo based methodology digitalhub.fifa.com/m/f99da4f732...

right now - my model doesn't show playoff predictions but i think I will be able to do that going forward if i get some time.

but hey at least I have the right outcomes in here now