Developer Advocate at MotherDuck! Views expressed are my own and not my employer's.
Chandler, AZ
Joined July 2018
- Tweets5.7K
- Following854
- Followers3.8K
- Likes18.5K
Pinned Tweet
You have to really love something to write a book about it. DuckLake makes PB-scale lakehouses simple: Postgres, Object Storage, DuckDB, & 6 lines of SQL.
Chapter 1 of the @OReillyMedia book I'm co-authoring just dropped - download for free at this link!
motherduck.com/lp/ducklake-l…
Alex Monahan retweeted
I personally love doing maintenance on a DuckLake. Out of the box you get all the functionality you need to automate that part of a production-grade DuckLake. Amazing stuff.
thefulldatastack.substack.co…
Head on over to the top of the HackerNewz to hear about Why DuckDB 2.0 is Faster!
news.ycombinator.com/
Do you work in semiconductors like I used to? Increase your design velocity and save licensing costs for design software with custom data apps!
Dashboard requests @AheadComputing used to take 3 months (2,160 hrs).
With Dives, they take about 1.5 hrs (0.002 months). That's a lot less!
AheadComputing is building a high-performance RISC-V CPU from scratch. Until recently, an engineer who needed a dashboard had to spec it, wait for the design enablement team to build it, review it, revise it, and wait again.
Now hardware engineers make their own Dives: interactive data apps they describe to Claude Code, built on their data in MotherDuck. The design enablement team liked the side effect: "We removed ourselves from being the bottleneck."
What else changed with Dives:
🦆 A Claude skill deploys each new Dive to the team's internal production site
🦆 One power engineer updates his Dives during meetings and has an updated dashboard within the hour
🦆 Dives of test run schedules and license usage exposed the bottlenecks that cut CI wait times from 8 hours to 1-2
🦆 Every Dive across every team is linked in one catalog and shared in weekly design reviews
But the big one was this:
"Every meeting now is shifted from trying to find answers to what to do about them."
Could be you: motherduck.com/case-studies/…
I gave my take on the topics coming to Data Outpost! AI and data are linked now, just like software and AI. It's up to us to learn how this new world works, and a lot of the experts will be at the Outpost to point us in the right direction.
motherduck.com/blog/data-out…
Alex Monahan retweeted
Virtual Workshop: Build a Data Agent in 60 Minutes xitter.cf/i/broadcasts/1XxygwNyb…
Alex Monahan retweeted
We Classified 100,000 Rows in 40 Seconds: Introducing prompt_jev() xitter.cf/i/broadcasts/1dGYlaOyy…
Alex Monahan retweeted
One week into shipping direct Jev support in MotherDuck SQL and we have (1) maaaaaany billions of tokens generated and (2) way more customers running real production analytics pipelines using prompt_jev() than I expected!
What are real customers using prompt_jev() for?
- The top use cases are tagging product and event data of varying shapes, and labeling and scoring support tickets, free-form text, and of course sales calls.
- Agent productivity is more common than I expected. Lots of customers have tons of agent traces and want to tag them by what they're about. Everyone's trying to get to the bottom of dev productivity & new product analytics w/ their own agents.
- Customers of all sizes and stages use it. Big, established accounts, small nearly-dormant ones who don't want to talk to sales springing to life, everything in-between, and tons of new signups who upgraded to our Business plan and immediately started cranking.
I'm going live tomorrow (Tues Sept. 29th) at 9:30am PT with @matsonj to dig deeper. Link in reply
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!
Alex Monahan retweeted
I've been researching and writing about DuckDB for over a year now. I've decided to lay out all of my content into a single page learning path. More will be added and maybe one day I'll turn it into it's own website and platform.
thefulldatastack.substack.co…
Alex Monahan retweeted
I decided to take the research and expert consensus on what makes a data visualization effective and turn it into an open source agent skill that anyone can use.
It's called evident-charts and works in Claude Code, Codex, Cursor and more. Really happy with how it turned out!
Alex Monahan retweeted
𝟭𝟬𝟬,𝟬𝟬𝟬 𝗿𝗼𝘄𝘀. 𝟰𝟬 𝘀𝗲𝗰𝗼𝗻𝗱𝘀. 𝟴𝟵% 𝗮𝗰𝗰𝘂𝗿𝗮𝗰𝘆. $𝟬.𝟱𝟬.
Next Tuesday, we’re going live to demo our new prompt_jev() function, which brings TypeSafe AI’s Jev model to MotherDuck for fast, SQL-native text classification.
Jacob Matson, Hamilton Ulmer, and Dumky de Wilde will run it against a live dataset, explain what makes Jev different from a general-purpose LLM, and unpack the benchmark
September 29, 9:30am PDT
Save your spot: luma.com/prompt_jev
As a data viz nerd, I had to get dbt Charts running on MotherDuck. 7 days later, I had a dbt package: custom SQL macros, serverless Python scripts, TypeScript plotting libraries, all in 2 configs.
You can build just about anything on MotherDuck. This job is just so much fun!
dbt Labs launched Charts last week at #dbtSummit and you can already publish them as MotherDuck Dives! Two config changes then pipelines and dashboards ship in one PR.
motherduck.com/blog/dbt-Char…
Alex Monahan retweeted
Last Wednesday afternoon, moments after I got access, I vibe-coded a Jev/DuckDB extension
Five days later, customers are paying to run Jev on their tables in MotherDuck via SQL
Startups rule
Alex Monahan 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!