Sabitlenmiş Tweet
My summary of lessons learned applying data science.
English
Vincent D. Warmerdam
12.1K posts
Vincent D. Warmerdam
@fishnets88
Prefer common sense over hype. Employed @marimo_io, built https://t.co/KOwpfLMpuQ blog https://t.co/zREZ8H2IBZ.
Amsterdam Katılım Haziran 2011
2.2K Takip Edilen11.4K Takipçiler
@vercel we maintain a bunch of skills over at @marimo_io for our users.
What's the process for getting on the "official" list?
English
English
@mitchellh Thing that works for me: write down the stuff that needs to get done today on a piece of paper.
That piece of paper is your manager for the day.
Crucial detail: at the end of each day you must also write down a thing that you learned. Got nothing? You just wasted a day.
English
This is how you get dumber btw, true even before AI. Turn on DnD, put your phone in a drawer. The best option if you can is to separate work + personal devices so your work device can't even see personal stuff. This is also partially why Pomodoro was all the rage a decade+ ago.
𝐑.𝐎.𝐊 👑@r0ktech
POV: you’re a developer in 2026😂
English
Related: if y'all are open to it, I made a PR to mention the marimo ACP on the opencode docs. Too many people still don't know that opencode is supported.
github.com/anomalyco/open…
English
English
It’s still alpha, but already useful in local dev workflows.
If this sounds useful, I’d love people to try it and tell me what works and what doesn’t.
And @fishnets88, curious whether there’s a marimo-shaped version of this idea too?
English
This resonated enough that I went and built my version of it: agentnb.
It gives coding agents a persistent Python REPL they can drive through simple CLI commands, so they can iterate in-memory instead of constantly bouncing through scripts and restarts.
@lateinteraction
Omar Khattab@lateinteraction
Though bash is a completely valid REPL, the amount of time coding agents lose during experimentation because they iterate on scripts instead of a Jupyter-like in-memory REPL is basically dumb. Fixing 1 local bug should not require restarting the whole job. Need better scaffolds.
English
By the time you see this screenshot, the feature will be live!
Think of it as a new kind of blog where we celebrate Python notebooks :)
... and widgets! So many novel widgets!
English
Vincent D. Warmerdam retweetledi
mo motebooks fewer problems
Hasan@hhyurdagul
For a long time, every data science or machine learning project had a folder called notebooks in the root directory, reserved for Jupyter notebooks that render nicely on GitHub. I haven't used Jupyter for over a year; instead, I've been using @marimo_io. I realized that when people see a notebooks folder in a GitHub repo, they expect to find rendered notebooks rather than just static Python files. Since marimo isn't widely known yet, people may not immediately understand that the project uses it. So I came up with the name Motebooks "Notebooks, but for marimo", so it's instantly clear that it's something different from Jupyter and worth checking out.
English
This was a fun rabbit hole indeed!
We recently added gcc to molab containers for Cython support, but it turns out that technically ... you could also use Zig! It is available via pypi!
That's an interesting unlock! Feel free to read the nerdsnipe here:
cemrehancavdar.com/2026/03/05/zig…
English
Shoutout to @askalphaxiv, their stack makes it super fun to explore papers!
English
What's the weirdest prompting trick you know?
Is it repeating the same prompt twice?
No. It's repeating the prompt ... three times.
Details and benchmark in this video: youtu.be/j2viZGPSydY
YouTube
English
Notebook productivity is going only up now that it is easy to make your own library of JS <-> Python widgets to help you do your work. The interaction is *super* helpful.
English
Vincent D. Warmerdam retweetledi
Makes me happy to see so many people build cool shit on pi :)
English
@HamelHusain What makes it feel better than say conductor?
I'm giving it a spin now and not I'm sure I "get it" yet.
English
Codex Desktop app is the first interface that has taken me away from the terminal. It's that good
English
@fatih What makes 2U more ergonomic?
I might need to start going down that QMK rabbit hole myself ...
English
@fishnets88 For ergonomic reasons. I want my Enter or Space buttons to be big. Even normal non-ergo keyboards have them at least in 2U. Kinesis got it right. I wish others would follow their leads.
I use QMK always and flash my keyboards.
English
Finally, someone from our Mechanical Keyboard community (Thank you Özkan) lent me a hand. I wanted to change the switches of my Kinesis 360 for a long time. Unfortunately, they are not hot swappable, so I need to desolder them. He desoldered the existing switches of my Kinesis 360 and is now changing them with better ones. The new switches are Gateron Beer (tactile), and the previous ones were Kailh White Clicky. I never liked them because of the feel. The Gateron's will be much better. Looking forward to using them.
English
I'm recording a video for all my fellow bayesians to help explain why a break even point is less relevant than the distribution of the break even point.
In the same video, I also explain how to code defensively so that Claude actually picks the correct distribution.
English
@fatih Interesting ... is there a reason why you want 2U?
I love both Elora and Imprint, but VIAL does fail me sometimes.
English
@fishnets88 They are better than the builtin Kahl Clicky ones. Imprimt is cool, but has the same issue as mine Elora. The thumb keys are all 1U. I want 2U.
English