Eliana is a backend developer interested in data systems, looking for work based in NYC. She's a Recurse Center alumnus and has been an active member of NYC Systems and the Software Internals book club. She's smart and thoughtful. Talk to Eliana. RT for reach. www.linkedin.com/in/eliana-ab...

From our first days in starting @b12.io, we've enabled customers to describe what they wanted and have us worry about how to get it done. With the AI Agent, this can happen near instantaneously. Really proud of the team for getting us here!

I've been thinking about this all day

[new blog post] Supporting our AI overlords: Redesigning data systems to be Agent-first muratbuffalo.blogspot.com/2025/09/supp...

Not something you see every day: while looking at a database cluster, I noticed that the "follower" machine had higher CPU utilization than the "primary." At some point, without us (and our customers) knowing, the primary failed over to the follower and was replaced. What a world.

I appreciate @werd.io's notion of a career mission (his is "to work on and support projects with the potential to make the world more informed and equal.") as a way to guide one's technical skills toward positive change. werd.io/using-techno... (h/t Mike Olson)

Where is the Cursor for SQL? CoQuery is a minimal vibe-coded SQL playground powered by SQLite WASM and Google Gemini language models. There is no backend, everything is in the webpage frontend. github.com/arnabdotorg/...

There's almost always a surprising nugget, usually multiple, in the acknowledgements section of a book/dissertation. Here's one from Manning Marable's "Malcolm X: A Life of Reinvention," for which he was posthumously awarded a Pulitzer prize in history.

Congratulations to @hmason.bsky.social and team! What a nice way to balance creative expertise and machine generation!

Olivia Rodrigo performs with Weezer at Lollapalooza www.punknews.org/article/8620...

Excited for this book. From @beenwrekt.bsky.social's blog post: "We decided people should make decisions like computers...The problem is, this decision was irrational. We can only optimize simple problems. Parlor games don’t describe how people behave."

Build your website directly from Claude with the B12 Website Generator!

🚀 We've just open-sourced Embedding Atlas – a tool for exploring large embedding spaces through rich, interactive visualizations 📊.

From some database performance legends, too!

ICML found hidden prompts in accepted papers. They have released a statement icml.cc/Conferences/... Yes, it’s unacceptable. So is using an LLM to review a paper. Peer review is so broken.

One of the great things about decentralized social media is that researchers can experiment with new ideas. Please help science and apply to be part of this study that one of our students is doing to make it easier for anyone to create sophisticated custom Bluesky feeds. Spread the word too! 🙏

Just found out "vibe-coding" is not "looking at a program and assigning it a natural number based on its chakras" wtf

The weekend is here! Perfect time to submit your #dataBS talk. We want your lessons-learned stories from data work, everything from "my data pipeline: the unsung hero" to "how we got that AI system working" to "what we learned when it fell over." Details and sign-up: bit.ly/dataBSconf-cfs

As more engineers use coding agents professionally, the traditional junior engineering mentorship model delays two skills they need to be successful with agents. Junior engineers need earlier introductions to task definition and code review. I wrote about that here: blog.marcua.net/2025/07/21/m...

As more engineers use coding agents professionally, the traditional junior engineering mentorship model delays two skills they need to be successful with agents. Junior engineers need earlier introductions to task definition and code review. I wrote about that here: blog.marcua.net/2025/07/21/m...

Working on a blog post on "Junior mentorship in the age of coding agents" in case anyone wants to provide early feedback!

Based on a fun discussion here last night, I wrote about that AI coding study and why quantitative “measurements” in social science are always ethnographic vibes.

Though much remains uncertain (e.g. outcomes of lawsuits re: canceled grants), it is clear that U.S. gov funding for science — especially science that intersects w/ concerns about human flourishing — is going to be significantly diminished in coming years. Here are a few things we can do to help:

Happy early Q3 kickoff meeting to all those that celebrate

@avibryant.com is a standup guy macleans.ca/politics/tax...

New post: Today I’m releasing a new Python command-line library that allows users to ask questions of federal campaign finance filings by leveraging large language models, with some built-in help. thescoop.org/archives/202...

I work in the best university in the world, less than 3 hours from my rural hometown, yet superstar students travel here from all over the world… if we keep treating them badly, will centers of excellence move elsewhere? "If my kids excel, will they move away?" jeffreybigham.com/blog/2025/wh...

NYCers, if you aren't registered to vote, the deadline is today!! e-register.vote.nyc?ref=hellgate...

New longform @inkandswitch.com essay! 📜 Malleable Software: Restoring user agency in a world of locked-down apps by me, @joshuahhh.com, @pvh.ca and @seaofclouds.com inkandswitch.com/essay/mallea... It's about why people need agency over their software tools, and how to make that happen.

Things you can learn about a place from Google's automatically generated summary... (Also Less Than Jake is playing there June 17 if you're into that)

Very elegant. Supports multiple compute nodes, time-travel, and even views with metadata in any boring relational DB. Assuming it supports multiple writers, this is great! I'm curious how compaction works/interacts with versioning/time-travel. Congratulations @hannes.muehleisen.org and team! :)

I made a mini-tool called Rich to enrich your CSV files with additional columns using the OpenAI API. It runs entirely in the browser, and you can use it here: marcua.net/minitools/ri... Blog post here: blog.marcua.net/2025/05/22/r...

Eager to contribute to democratizing insights from tabular data? We have 2 new PhD openings! ✨ 1) Fundamental Techniques in Table Representation Learning 2) Reliable AI-powered Tabular Data Analysis Systems ⏰ Apply by: 30 June 2025 📅 Start: Fall/Winter 2025 🔗 Info: trl-lab.github.io/open-positions

I made a mini-tool called Rich to enrich your CSV files with additional columns using the OpenAI API. It runs entirely in the browser, and you can use it here: marcua.net/minitools/ri... Blog post here: blog.marcua.net/2025/05/22/r...

There were layoffs at MS yesterday and 3 #Python core devs from the Faster CPython team were caught in them. If you know of any jobs, please send them their way: Eric Snow: www.linkedin.com/in/ericsnowc... Irit Katriel: www.linkedin.com/in/irit-katr... Mark Shannon: www.linkedin.com/in/mark-shan...

Wrote up some notes about eng levels and AI skamille.medium.com/things-i-cur...

Grist for the AI mill skamille.medium.com/some-tips-fo...

JUST IN: ProPublica’s “Life of the Mother” series, which exposes the fatal consequences of abortion bans, has won the Pulitzer Prize for Public Service. This is the second consecutive year we’ve been awarded this distinction and our eighth Pulitzer: www.propublica.org/series/life-...

How does using GenAI tools reshape knowledge workers’ critical thinking? Our #CHI2025 paper studied 319 knowledge workers to dive into this question. w/@advaitsarkar.bsky.social @levlevlev.bsky.social Ian, Sean, Richard, Nick @msftresearch.bsky.social @hcii.cmu.edu www.microsoft.com/en-us/resear...

US students. Looking to apply to graduate school? Had an offer rescinded recently? The University of British Columbia in Canada will re-open the application portal in some departments for US students from April 14-18. Details: www.grad.ubc.ca/us-applicant... #ubc #AcademicSky #PhDSky #ScienceSky

Aside from being a bit too opinionated about Corollas, this is a nice blog post

It turns out to be hard to evaluate natural language with natural language. What should we take away from the conundrum of LLM evaluation? www.argmin.net/p/evaluation...

I found two GiveWell podcast episodes on USAID funding cuts to be helpful in understanding how the dust is settling (not that it's fully settled): - Late March on the big picture: blog.givewell.org/2025/03/20/p... - Early April zooming in on a specific type of aid: blog.givewell.org/2025/04/03/p...

Prompt injection, but on your machine, seems like the scariest of these!

Aditya and I had fun speaking with Ben's Bites and demoing what B12 has built. Launch-ready website in 60 seconds, and all the goodies to iterate from there!

The Office of the New York State Attorney General is hiring a Director of Artificial Intelligence: www.linkedin.com/jobs/view/di... ...this seems like the kind of role that needs someone great. Is it you?

not okay

Hard to believe this has only existed for 10 years! It was such a helpful resource for us when we established our ladders and rubric. ...and how refreshing to read "maybe you don’t need to create a job ladder for literally every job"