Andrea Lattuada

Systems software and first order logic. PhD from @SystemsGroupETH, then @vmwareresearch, MPI-SWS. Now eng @turbopuffer

[Simon Eskildsen @Sirupsen]

An incredible year for turbopuffer is closing. We now manage trillions of vectors and tens of petabytes. We’ve puffed up revenue 10x and headcount 5x.

We’ve helped Anthropic, Cursor, Notion and others connect every petabyte to AI to ship the most ambitious version of their product.

This week we doubled down with Lachy Groom and added Thrive Capital to the team. They have both proven themselves repeatedly to us over the past 2 years.

In 2025, we’ve proved the puffer architecture is capable of state-of-the-art vector and text search. Next year, we’ll continue to optimize the core and widen our surface area, to make it effortless to search every byte.

[Andrea Lattuada @utaal]

Papers We Love is back in Zürich! Join us for the first event of this new series on Thursday November 20th at ETH: Abhi will talk about the elegance and magic of Automatic Differentiation of programs (and how it's a generalization of backprop). https://t.co/nBu7X2FVpe

[MPI-SWS Operating Systems Research Group @OSGroupSWS]

Today, Johanna Polzin from @CSatETH gave a talk on her ongoing work with Matthias Brun, @achreto , and @utaal , "Specifying and Verifying Page Tables in a concurrent OS"!

[Tianyin Xu @tianyin_xu]

Xudong (@xu_dong_sun) will soon be on the job market. He is very determined to and well prepared for fundamental systems research on software correctness and reliability. If your department is hiring, give him a chance to show you his work and future plans.

[Andrea Lattuada @utaal]

@frankmcsherry Thank you Frank, also for being an inspiration at the start of my academic career, and for teaching me how to do impactful, rigorous research.

[Andrea Lattuada @utaal]

@tianyin_xu @mpi_sws_ Tianyin, that’s incredibly kind, thank you!

It’s been such a pleasure to work with you, @xu_dong_sun, Wenjie, Tyler, and Zicheng on Anvil and I hope to continue collaborating with you and your group: it’s a dream team!

[Andrea Lattuada @utaal]

@vaastav05 @HerrDreyer @mpi_sws_ Thank you Vaastav! Happy to be starting soon!

[Andrea Lattuada @utaal]

@Mh_khoshechin @HerrDreyer @mpi_sws_ Thank you Mohammad! I’m looking forward to many interactions, discussions, and collaborations with all the fantastic folks at MPI-SWS.

[Andrea Lattuada @utaal]

@bitshiftmask @HerrDreyer @mpi_sws_ Thank you James! We should catch up some time. I’ll send you a PM.

[Derek Dreyer @HerrDreyer]

Two best paper awards at OSDI for projects using the Rust verification tool Verus!

Co-leader of the Verus project Andrea Lattuada (@utaal) will be joining @mpi_sws_ as a research group leader in September, and is actively looking for strong students & postdocs to work on Verus!

[Andrea Lattuada @utaal]

@xu_dong_sun @tylergu_jiawei @tchajed @_lalithsuresh @tianyin_xu Thank you @xu_dong_sun for leading this project with such energy and commitment!

[Xudong Sun @xu_dong_sun]

Shout out to my AWESOME colleagues Wenjie Ma, @tylergu_jiawei , Zicheng Ma, @tchajed , Jon Howell, @utaal , Oded Padon, @_lalithsuresh , Adriana Szekeres, and @tianyin_xu !

[Andrea Lattuada @utaal]

@xldenis whoa that was a blast from the past

[Emily @the_aiju]

you’ve heard of permaban, shadowban, etc, what if there’s a type of ban that makes all your messages show up at least 30 min later. i call it deutsche ban.

[Randall Munroe @xkcd]

d65536 https://t.co/LEmYO7AkEh

[Andrea Lattuada @utaal]

@m_ou_se Got it, thanks!

[Andrea Lattuada @utaal]

@m_ou_se So, uhm, how are you "reading the compiler errors in the future"? (could you share the code?) I promise I have no intention of actually using this as part of a rust formal verification project (maybe) :)

[@jxf.me (Bluesky) / [email protected] (Mastodon) @jxxf]

Optimist: The glass is ½ full.
Pessimist: The glass is ½ empty.
Excel: The glass is January 2nd.

[Alexis King @lexi_lambda]

it is a common misconception that the name “C++” is a joke implying that C++ is “one better than C”. in fact it implies that C++’s existence makes C seem better in comparison, due to its use of the post-increment operator.