@ben_lern
Account based inUnited States
About this account
- Account based in
- United States
- Connected via
- United States App Store
Account-level information from X, not a live location or the device used for a specific post.
cofounder @ Espresso AI - we use AI to optimize your snowflake bills
New York, NY
Joined February 2010
- Tweets186
- Following212
- Followers537
- Likes51
Don't be a meat proxy - be a meat cheerleader.
Even AGI benefits from encouragement:
- "Keep going!"
- "This is great!"
- "Take a break and get back to it, you can do it."
and my personal favorite:
- "Your big brother solved Navier-Stokes. I believe in you!"
Ever since undergrad you’ve wondered: “what is graph theory good for?”
Today, I’m excited to present the answer live at dbt summit - come learn how we use scheduling to run your dbt workloads faster and cheaper.
I’m talking alongside Saravana Kumar, a dbt user and one of our customers, who’ll tell you about how Authentic uses dbt, Snowflake and Espresso in their stack.
This makes debugging a ton faster than pulling logs manually - sometimes even faster than asking claude.
You can check it out on our free observability suite: espresso.ai/post/observabili…
Huge thank you to Andy at @babylist for sitting down with @Snowflake to talk about how they use the platform - and how Espresso AI helps.
(They got 20% of their data engineering time back from tedious manual optimization while dropping costs at the same time - exactly the sort of outcome we hope for for our customers.)
Check out the video: youtube.com/watch?v=DuChT5Jn…
happy labor day - the one day of the year you come in to work for free.
honestly I was never sure why they made this a holiday
You can use graph theory to cut dbt runtime without buying more compute!
dbt builds a DAG, then just routes all your queries to one warehouse anyway. You can do better by parallelizing queries with no dependencies. We've tried this, and in production customers have seen dbt jobs run up to 50% faster with no increase in cost.
Read more at espresso.ai/post/dbt-acceler…; I'll also be talking about this at dbt Summit if you want to hear more detail live :)
We're hosting a dinner during dbt Summit this year and calling it "Bill Therapy".
Data warehousing costs are nuts, so we're getting some data leaders in a room to compare notes (and maybe try some nice wine).
If you're at dbt and your Snowflake or Databricks bill has become a line item you're constantly fighting with, you'll be in good company.
It's happening at Scarpetta at the Cosmopolitan on Wednesday, 9/16.
Seats are limited, but RSVP soon and I'll approve you
luma.com/h3j1jn82
Our #1 rule for sales reps running pilots: Ask not what Engineering can do for you, but what you can do for Engineering.
🤖 Made with AI
- Managing your own nerves and giving engineers space to work. You don't have control over the tech, so it's tempting to ask for updates - but you don't need two 30-minute debriefs a day, and the engineers don't have an hour to spare to hang out.
