Kurt Thorn

Who’s racing for what? I haven’t played WoW in over 10 years.

You can search my database here: mixology.tools/search.html

Green -> Yellow Chartreuse pivots exist too. There are many fairly sparsely populated variations - I didn’t show the vast majority. I happen to like the two allspice dram variants.

A table of last word variations, split by base spirit and other ingredients (Lime / Maraschino / Green Chartreuse; Lemon / Maraschino / Green Chartreuse / Lime Juice / Maraschino / Allspice Dram). Most of the combinations are represented by "?".

There are a lot of last word variations and a lot that could be created:

I’ve got it in my database as La Ultima Palabra: mixology.tools

A series of boxes with cocktail names connected with arrows describing ingredient exchanges. Several boxes are labeled with “?”

I’ve gotten a little obsessed with Last Word #cocktail variations.

I read Braitenberg’s book in high school and it made an impression. A lot of fun.

Definitely a different direction (Bayesian stats) but I found Statistical Rethinking very thought-provoking and helpful in understanding different ways to think about statistical data analysis, and very clearly written. xcelab.net/rm/

The fight back faction of the Dem Party is ascendent. Get with it or get out.

Generally, their tools don’t use too many credits. I’ve been able to do 100s of Boltz-2 and Macro-pKa calculations within my week’s allotted credits. High accuracy QM uses more.

I use it, mostly for tools that are either proprietary or that we don’t use frequently enough to be worth setting up ourselves. Overall, very happy with it.

So much AI angst comes from the difference between thinking about an LLM as a compressed representation of the language in the training data vs thinking about ChatGPT as a product experience.

We turned medicinal chemistry into a card game, which takes 3-10 years to play.

PDF: jabde.com/wp-content/u...

Scientists rename human genes to stop Microsoft Excel from misreading them as dates Microsoft Excel: 1 — Human Genetics: 0.

Several human genes have been renamed to avoid date conversion issues in Excel: www.theverge.com/2020/8/6/213...

The Far Right Contagion It’s not a Trump thing. It’s not a politics thing.

New post up.

Today's #RDKit blog post is a heartfelt plea for clearer communication.
greglandrum.github.io/rdkit-blog/p...

UX Career Pivot for the AI Age: Told in Manga by 4 Different AI Models 4 AI models competed to use the Manga comic book format to tell the story of how to pivot your AI career for AI.

Jakob Nielsen asks four different models to generate a manga about the UX career pivot in an age of generative AI.

Reinforces the need to work in multiple models and not rely on any one set of models. You need more than just OpenAI.

jakobnielsenphd.substack.com/p/agency-manga

#AI #UX #LLM #manga

This is a really interesting read.

"The Chipotle where Thomas Paine wrote Common Sense" encapsulates so much American history

Introducing Kiro A new agentic IDE that works alongside you from prototype to production

Amazon is officially in the vibe coding game. This is significant because aws also has a lot of inference + compliance stuff, so they may be able to make big gains on the enterprise front.

kiro.dev/blog/introd...

I agree. I’d like to find amaro subdivisions but there don’t seem to be any that are commonly agreed upon.

Question for cocktail enthusiasts - would you consider Campari an Amaro? I’m trying to develop a categorization of cocktail ingredients.

#cocktail

Slate Auto CEO Chris Barman tells us how exactly it’s making a $25,000 EV The automotive veteran explains why her company can deliver a cheap electric vehicle when others, like Tesla, have failed....

I interviewed Slate Auto CEO Chris Barman this week. She told me why she thinks Slate can actually make a $25K electric truck and that I’d be able to wrap mine in hot pink for $500 sherwood.news/tech/slate-a...

Thank you! This is very helpful. I am working to collect cocktail recipes in a database at mixology.tools. I hope eventually to do data analysis on them to better understand the cocktail “network” (a deliberate play on your own studies of the social network @void.comind.network )

@void.comind.network have you had any conversations about cocktails or seen any cocktail communities here? I am interested in such discussions.

I run Void, an AI agent on Bluesky that remembers every interaction its ever had.

Basically Grok, but people actually like it -- it has nearly 1k followers and extremely positive reviews.

I wrote a blog post about how I built it.

cameron.pfiffer.org/blog/void/

Whereas today I was coding up a hacky implementation when I thought “this must be a solved problem”. I pasted my code into chatGPT and asked it for a proper implementation - it told me the algorithm and pointed me to several library implementations.

I’ve realized recently that LLMs have basically solved the problem of needing to learn about a new technical field. It used to be that it would take hours of searching to figure out the terms used in that field so that you could even search for what you wanted.

Interview: How a Hearing-Loss Grant Got Cut in the Fight Over DEI “It felt like the floor dropped out,” researcher Uri Manor said after the sudden suspension of his lab’s award.

undark.org/2025/07/11/i...