-5 days to the DEEM Workshop @ SIGMOD at ACM SIGMOD/PODS! We will present "Reproducible #data science over data lakes", sharing our progress in combining Dremio's Nessie with a multi-language cloud runtime. Pre-print here: arxiv.org/pdf/2404.13682! See you in #chile, data cowboys!

Two DBOS decorators, one bauplan import, and you're off to the races: write-audit-publish pattern in pure #Python - no JVM, Spark or Iceberg-specific API to learn, a lakehouse that just works. Working with Peter Kraft and Qian Li was a blast: thanks for your help!

Warehouses and notebooks are great, but they won't solve your AI issues: data pipelines, CI/CD, private packages, reproducibility. The why-and-how of our “programmable lakehouse” (FaaS + GitHub-for-data) is now out for the first time: thanks Kaivalya Apte - The Geek Narrator!

“You could attach prices to thoughts. And how does one pay for thoughts? The answer, I think, is: with courage.” Can we re-imagine the data lifecycle so that it works for humans and machines alike? Yes, we can. Thanks Davis Treybig (and team) and Aditya Agarwal (and team)!

True, but it doesn’t have to be! pip install bauplan bauplan create-table mybucket mytablename

Even as second time founders, South Park Commons has been among the very best choices we ever did in our startup life. Proud to be a South Park Commons founder, even more so today: congrats Aditya Agarwal and team!

When Paul Dix is in the house, expect unmatched data wisdom and incredible shirts, not necessarily in this order!

Thanks Qian Li! Come for the food, stay for a frenetic journey in databases, git-for-data, fast containers and many more things I don’t know anything about!

A fun way to learn #alloy for distributed systems led us down the rabbit hole of guiding a struggling #opus with formal verification. From math Olympics to a magazine puzzle, surprises abound in the finicky #llm world. On the plus side, logic in #ai is cool again 😎

“Make something idiot proof and somebody will come up with a better idiot”: in a world of #llm, data platforms built on Git-for-everything and simple APIs may offer a new foundation. Thanks Nitay Joffe and Kostas Pardalis for the fun chat!

💯That’s because Federico Bianchi is a genius and I knew it before most 😎