Pinned Tweet
Is it finally time for a new query language to replace SQL?
I was curious what @lloydtabb and his team were doing with Malloy, so I took some time this long weekend to check it out. The short version is: I'm super excited.
The long version is here: carlineng.com/?postid=malloy…
Carlin Eng retweeted
Replying to @BachFrancis
(1/7) I'd like to provide some additional context for the remarks that can be found in this interview.
(2/7) My discomfort with the attention being paid to existential risk---and with the ongoing dominance of the public discussion of such risks by a few so-called thought leaders---is that there's an elephant in the room that is being missed. That elephant is the effect of all of this discussion on young people, and indeed the effect---beginning with social media a decade ago---of information technology on young lives broadly speaking.
(3/7) I recently read an interview by Jill Lepore in which she reports on discussions that she had recently with a group of college students: "They had just three things to say. One, social media completely destroyed their childhoods. Two, AI is destroying their college education and their likely prospects for employment. And three, what the hell are we all doing?" To add a fourth bullet that I've heard in similar discussions: "I'm not going to have children, because the existential risk is just too high."
(4/7) Please contemplate such a statement, those of you who think that it's OK that the public, widely diffused, discussion coming from seemingly authoritative figures is so adamant about existential risk.
(5/7) Let me add that I have little interest in a debate about what "superintelligence" might mean, or whether we're currently seeing the emergence of superintelligence. For what it's worth, I would certainly agree that the solution of extremely challenging math problems (for example) exhibits a very impressive form of intelligence, although I would want to argue that such intelligence is impossible to dissociate from the collective of humans that provided data to the machine learning system (which itself has become a kind of collective).
(6/7) Such philosophical debates are entertaining and even important, but when they then head in the direction of "when such superintelligence emerges, there's a good chance that it will want to kill all humans", I say "science fiction". Again, I don't think that it's bad to consider such thoughts, but I strongly believe that it's bad for society, and especially young people, for such thoughts, unaccompanied by any serious real-world analysis, to become a dominant form of discourse of the era.
(7/7) We are damaging human lives, and the consequences of that damage will no doubt be a major theme of the next several decades.
Carlin Eng retweeted
if you value intelligence above all other human qualities, you’re gonna have a bad time
Carlin Eng retweeted
THIS IS MY FAVORITE USE CASE SO FAR
A question we've been asking for years is:
what happens when intelligence/$ is so good you can apply to every row in your database or every line in your log file
Text classification in MotherDuck just got ~50x faster at ~1% of the cost.
prompt_jev() is a SQL function powered by Jev, TypeSafe's new system one model. 100k rows: 40s, $0.50, frontier-LLM accuracy. The LLM took 32 min and $37.
Read on:
motherduck.com/blog/motherdu…
Carlin Eng retweeted
Text classification in MotherDuck just got ~50x faster at ~1% of the cost.
prompt_jev() is a SQL function powered by Jev, TypeSafe's new system one model. 100k rows: 40s, $0.50, frontier-LLM accuracy. The LLM took 32 min and $37.
Read on:
motherduck.com/blog/motherdu…
Carlin Eng retweeted
Today we're launching prompt_jev(), which brings Jev, a new kind of AI model from @typesafeai, to MotherDuck SQL.
Jev makes text classification super fast, dirt cheap, and as easy as prompting an LLM. In our tests, it ran at 50x the speed and 1% the cost of comparable frontier models, with similar or better accuracy.
For analytics, this opens up scoring and labeling workflows that used to be too slow or costly. Classify all your data, rerun when your labels change, or filter on meaning in a WHERE clause while doing exploratory data analysis.
Live now in MotherDuck!
Carlin Eng retweeted
I made a DuckDB extension where you can use @typesafeai 's Jev to do quick classification of rows in any csv/parquet file or duckdb table
about 10sec for 1k rows ~ better than using an LLM, way more ergonomic than a classifier
game-changing for data analysis!
Carlin Eng retweeted
This is wrong. You don't have to analyze every possible legal game to solve a game. It's not even close.
Transpositions reduce the number of positions you have to look at. In chess there are only about 10^44 possible positions.
And more importantly, tree pruning reduces the required search tree much further. Simple alpha-beta pruning reduces required search vastly. And there are other ways to optimize the search.
For example, Gomoku has about 10^105 possible positions (much more than chess!), but was solved by Victor Allis by analyzing only about 5 million positions. And then the winning solution tree was reduced further to about 130,000 positions.
Carlin Eng retweeted
⌯⌲ Introducing Flights, a new agent-native data pipeline tool
Author in python / SQL, trigger in SQL + MCP + on a schedule. I use it for data pipelines, but also for agent workflows, classical ML tasks, screenshotting Dives and posting to Slack, you name it.
Give it a spin!
Carlin Eng retweeted
2026 is wild. Any data viz you want is just a prompt away... I wrote a post about using Claude Code to build Dives - super interactive data visuals! To test the limits, I made a premium-level pivot table from scratch. See how to build Dives of your own!
motherduck.com/blog/claude-c…
Carlin Eng retweeted
Replying to @Afinetheorem
I've done quite a bit of watching of Bending Spoons, so I can try to talk a bit about how they try to structure their company and their finances and stuff like that.
- I would say "neglected tech companies" rather than "failed tech companies" because they have a bunch of users and revenue.
- Their key insight is that you can run these companies and make a lot of money if you cut the staff by 90% and increase prices by 90%.
- You don't need thousands of engineers to run many of these tech companies. Thousands are often needed to build up to what they have today, but once the core infrastructure is in place and in maintenance mode, fewer engineers are needed.
- If you focus on a couple of key factors, you can run these companies with a small team.
- They use tons of leverage and OPM to help with core equity economics.
- They spend a lot of time on engineering infrastructure and simplifying it, which helps reduce the core team size.
- This model works only if you have:
- A team that knows how to finance the deals
- A team that knows engineering well enough to know what to cut and what not to cut
- A really good engineering squad to run these massive companies with only a few hundred people.
Little value is created by transferring the company to an Italian entity. But they keep like 10% of the existing team and augment them with their internal team of a few hundred good engineers (think google mid-2000s quality). Then they can simplify, build out better infra so less maintenance is needed, raise prices, and expand.
Carlin Eng retweeted
introducing CSV Everything!
Transform tables and charts from images into downloadable CSV files. Simply copy any image containing tabular data to your clipboard and convert it to structured CSV format with one click.
Carlin Eng retweeted
.@abigale_kim's paper is unleashed! It's the most complete eval of DB extensions/plugins. We analyze @PostgreSQL, @MySQL, @mariadb , SQLite, @duckdb, @Redis.
TLDR: Postgres ecosystem is fraught w/ footguns. Other DBMSs have fewer extns but less problems. DuckDB has cleanest API.
Vol:18 No:6 → Anarchy in the Database: A Survey and Evaluation of Database Management System Extensibility vldb.org/pvldb/vol18/p1962-k…
"Have you tried rubbing a database on it?"
Today we're launching DuckLake, an integrated data lake and catalog format powered by SQL. DuckLake unlocks next-generation data warehousing where compute is local, consistency central, and storage scales till infinity. DuckLake is an open standard and we've implemented it in the "ducklake" DuckDB extension.
– Read the announcement blog post at duckdb.org/2025/05/27/duckla…
– Visit the new DuckLake website at ducklake.select/
Carlin Eng retweeted
The reality of building web apps in 2025 is that it's a bit like assembling IKEA furniture. There's no "full-stack" product with batteries included, you have to piece together and configure many individual services:
- frontend / backend (e.g. React, Next.js, APIs)
- hosting (cdn, https, domains, autoscaling)
- database
- authentication (custom, social logins)
- blob storage (file uploads, urls, cdn-backed)
- email
- payments
- background jobs
- analytics
- monitoring
- dev tools (CI/CD, staging)
- secrets
- ...
I'm relatively new to modern web dev and find the above a bit overwhelming, e.g. I'm embarrassed to share it took me ~3 hours the other day to create and configure a supabase with a vercel app and resolve a few errors. The second you stray just slightly from the "getting started" tutorial in the docs you're suddenly in the wilderness. It's not even code, it's... configurations, plumbing, orchestration, workflows, best practices. A lot of glory will go to whoever figures out how to make it accessible and "just work" out of the box, for both humans and, increasingly and especially, AIs.
Carlin Eng retweeted
Today's SQL or Death Seminar Speaker: @lloydtabb will present the @MalloyDev query language that seeks to replace SQL for data analysis. Zoom talk open to public at 4:30pm ET. YouTube video available after: db.cs.cmu.edu/events/sql-dea…
Carlin Eng retweeted
🎤 SPEAKER SPOTLIGHT: Data legend @lloydtabb (Looker, Malloy at Meta), presents "Building Blocks."
Learn from one of the industry's most brilliant minds at Data Council in Oakland, April 22-24.
Early bird ends 2/28! 🎟️ datacouncil.ai
Data Engineering lacks notions of reusability that other software engineering disciplines take for granted. I explore what I think is the root cause in my latest blog post. Spoiler: it's not about some subtle, vague difference. It's all about the tools. carlineng.com/?postid=holy-g…
See it live in action here 👀 youtube.com/watch?v=LlvjXfFV…
With Malloy’s semantic model, we can visualize query results like dashboards, tables, and charts with built in interactions shared across common dimensions and measures.
A preview of these new @MalloyDev features 🧵
Carlin Eng retweeted
With Malloy’s semantic model, we can visualize query results like dashboards, tables, and charts with built in interactions shared across common dimensions and measures.
A preview of these new @MalloyDev features 🧵



























