I’m really excited about this one! Building Krea’s data infra, I increasingly believe that AI research workloads are neither OLAP nor OLTP. This event has come out of some convos I’ve had with @willmanning, @ovaistariq, @HammadTime and others who have seen the shift happening. Even a small foundation model company in 2026 needs to be looking towards trillion row and exabyte scale. The huge GPU nodes you buy come with supercomputer levels of CPU to be used efficiently. GPU, memory, and storage prices are spiking. The tool landscape is moving towards disaggregated storage and streaming, and the scale is 10x’ing every year. All of the speakers are phenomenal distributed systems engineers, and I think it’ll be a lot of fun to have them all in one room. If you’re into databases and distributed systems, please join us!

Talked to dozens of frontier labs recently, and the biggest thing slowing their research teams down is clunky data pipelines that have to be refactored every time they want to try something new Come see the new data stack the best researchers rely on 👇

we're hosting a 'Big Data 3.0' next Tuesday (June 16) in our SF office with @SpiralDB and @TigrisData. we'll have technical deep-dive talks from frontier AI labs about internet-scale distributed data systems for AI research. details below 👇

Friends don't let friends train models with bad data

If you are interested in the intersection of AI x Science (neuro, bio), you're going to want to meet this team @CVPR tonight

Tonight's the night! AI is in its Science era Come join us if you're @CVPR in Denver this week!

We’re at @CVPR this week. Tomorrow evening, we’re co-hosting a happy hour with @SpiralDB for researchers, engineers, and builders working at the edge of AI for Science and multimodal modeling. Come say hi and meet the team.

FYSA: Spiral is SOC 2 Type II compliant 🔒

One of those charts you are excited to see go down 📉

We've been working on a major schema migration at @PolarSignalsIO that has unlocked some pretty wild improvements. Apart from a 50% (!!) reduction in storage size thanks to logical compression in @vortexdotdev, there has been a dramatic improvement in query performance.

Engineering at Spice AI, part 3: @vortexdotdev In late 2025, we selected Vortex as the premier columnar format for the Spice platform. It now powers Spice Cayenne in production. This post covers how we got there and what we learned: — What Vortex is and how it differs from Parquet — Why we chose it over alternatives — Cayenne's architecture: Vortex files SQLite/@tursodatabase metadata — Deletion vectors, compaction, & compression strategies — @ApacheDataFusio integration — 7 lessons from shipping it ➡️ hubs.ly/Q04b1l7L0

like i said, no one gets research into production faster than @SpiralDB

Initial PR here, currently feature gated while we iterate on perf & stabilize the encoding wire format github.com/vortex-data/vorte…