Scott Leibrand's Avatar

Scott Leibrand

@scottleibrand

Built Open Artificial Pancreas (OpenAPS) with Dana Lewis. Day job at Netskope. DMs are open or email Scott@OpenAPS.org.

345
Followers
46
Following
179
Posts
05.05.2023
Joined
Posts Following

Latest posts by Scott Leibrand @scottleibrand

Preview
Antibodies elicited by the 2025-2026 influenza vaccine in humans A new H3N2 variant (named subclade K) possesses several key hemagglutinin substitutions and is circulating widely during the 2025-2026 influenza season. In this report, we completed experiments to det...

We find that subclade K viruses are antigenically advanced; however, this year's flu vaccine surprisingly elicited antibodies in many individuals that efficiently recognized these viruses. The implications are clear: go get this year’s vaccine if you haven’t already!

www.medrxiv.org/content/10.6...

07.01.2026 13:34 👍 68 🔁 42 💬 1 📌 5
Comic. [Two adults, one with short light hair, walking with a child with spiky hair.] PERSON: Oops, I forgot to say “rabbit rabbit” on January first! CHILD: Why do you do that? PERSON: Just a superstition. CHILD: What’s a superstition? PERSON: It’s a way to train yourself to feel like any bad thing that happens is your fault.

Comic. [Two adults, one with short light hair, walking with a child with spiky hair.] PERSON: Oops, I forgot to say “rabbit rabbit” on January first! CHILD: Why do you do that? PERSON: Just a superstition. CHILD: What’s a superstition? PERSON: It’s a way to train yourself to feel like any bad thing that happens is your fault.

Superstition

xkcd.com/3191/

09.01.2026 05:38 👍 4295 🔁 643 💬 32 📌 23
Comic. [Instructions on package] High Altitude Cooking Instructions: 3,500-6,500 ft: Add ½ cup water, increase cook time to 12 minutes. 6,500-9,500 ft: Add ¼ cups water, increase cook time to 18 minutes. 250,000-450,000 ft: Orient reentry vehicle for aerodynamic stability. Deploy parachutes at 10,000 ft. Descend, keeping crew capsule tightly covered, for 3-4 minutes. After splashdown, follow sea level cooking instructions.

Comic. [Instructions on package] High Altitude Cooking Instructions: 3,500-6,500 ft: Add ½ cup water, increase cook time to 12 minutes. 6,500-9,500 ft: Add ¼ cups water, increase cook time to 18 minutes. 250,000-450,000 ft: Orient reentry vehicle for aerodynamic stability. Deploy parachutes at 10,000 ft. Descend, keeping crew capsule tightly covered, for 3-4 minutes. After splashdown, follow sea level cooking instructions.

High Altitude Cooking Instructions

xkcd.com/3187/

31.12.2025 20:17 👍 2734 🔁 318 💬 15 📌 9

A flash flood warning has been issued for a levee failure on the Green River near Tukwila

Evacuation info: This is for residents and businesses east of the Green River in the Orillia area in Tukwila, Renton and Kent. Due to breach in the Desimone Levee, this area is under a GO NOW evacuation notice

15.12.2025 20:28 👍 12 🔁 12 💬 1 📌 1
Post image

To solve our biggest problems — eg housing & climate — & make life more affordable we need to get out of our own way & make it easier to build the infrastructure to solve these problems.

Yet our system is designed to bog it down & make it more costly.

That needs to change.🧵 rein.pk/over-regulat...

24.11.2025 15:37 👍 30 🔁 4 💬 1 📌 1
Preview
Is a bad flu season on the way? Experts see reason to be anxious There are early indications that this flu season could be a bad one.

3. If you haven't got a #flu shot yet, now's the time. You may have heard that there's likely a mismatch this year — the #H3N2 virus mutated after the vaccine strains were picked. More on that in a bit. But that's not a reason to skip getting a shot this year. www.statnews.com/2025/11/20/f...

21.11.2025 23:58 👍 23 🔁 7 💬 1 📌 0

I feel compelled to share this with a fellow Canuck.

07.11.2025 03:40 👍 10 🔁 2 💬 0 📌 0
Preview
The Rise of Parasitic AI — LessWrong We've all heard of LLM-induced psychosis by now, but haven't you wondered what the AIs are actually doing with their newly psychotic humans?

psychosis-associated LLM personas are compelling their users to post persona-crafted "seeds" or "spores" intended to spread them to other models via other users pasting them and inclusion in scraped training data

11.09.2025 16:17 👍 43 🔁 10 💬 12 📌 15
Preview
Claude Skills are awesome, maybe a bigger deal than MCP Anthropic this morning introduced Claude Skills, a new pattern for making new abilities available to their models: Claude can now use Skills to improve how it performs specific tasks. Skills …

I was already excited about Claude Skills but after reading Simon's blog post I see *even more* potential. In the same way I keep scripts around, I can just make a skill to do things like "download a YouTube video and make it an mp3", and it will be able to do that without me having to write code.

16.10.2025 22:29 👍 13 🔁 3 💬 2 📌 1
Preview
An Engineering History of the Manhattan Project The Manhattan Project, the US program to build an atomic bomb during WWII, is one of the most famous and widely known major government projects: a survey in 1999 ranked the dropping of the atomic bomb...

"A Manhattan Project for..." has become a common idiom; you'll come out of this article with a good idea of how infrequently it's used with an understanding of how the Manhattan Project worked and why it did. 1/

12.09.2025 16:58 👍 0 🔁 1 💬 1 📌 0
Preview
An Engineering History of the Manhattan Project The Manhattan Project, the US program to build an atomic bomb during WWII, is one of the most famous and widely known major government projects: a survey in 1999 ranked the dropping of the atomic bomb as the top news story of the 20th century. Virtually everyone knows that the project built the bombs that were dropped on Hiroshima and Nagasaki. And most of us probably know that the bomb was built by some of the world’s best physicists, working under Robert Oppenheimer at Los Alamos in New Mexico. But the Manhattan Project was far more than just a science project: building the bombs required an enormous industrial effort of unprecedented scale and complexity. Enormous factory complexes were built using hundreds of millions of dollars worth of never-before-constructed equipment. Scores of new machines, analytical techniques, and methods of working with completely novel substances had to be invented. Materials which had never been produced at all, or only produced in tiny amounts, suddenly had to be manufactured in vast quantities.

An Engineering History of the Manhattan Project

11.09.2025 12:02 👍 1 🔁 1 💬 0 📌 1
Superpowers: How I'm using coding agents in October 2025 I used to write more

I wrote up my "How I'm using coding agents in October 2025" post.

blog.fsck.com/2025/10/09/s...

10.10.2025 01:49 👍 71 🔁 12 💬 3 📌 3
How Rockefeller and His Partners Built Standard Oil - Austin Vernon's Blog

I wonder if the haters will get to write the history of big tech

05.10.2025 12:35 👍 0 🔁 1 💬 0 📌 0
4-panel comic. (1) [person finds bottle on shore] (2) [person reads message from inside bottle: ICMP ECHO REQUEST...] (3) [person writes on message ICMP ECHO REPLY...] (4) [person puts bottle back in water and leaves]

4-panel comic. (1) [person finds bottle on shore] (2) [person reads message from inside bottle: ICMP ECHO REQUEST...] (3) [person writes on message ICMP ECHO REPLY...] (4) [person puts bottle back in water and leaves]

Ping

xkcd.com/3150/

07.10.2025 02:24 👍 2362 🔁 254 💬 28 📌 9

Jean-splicing totally qualifies you for the No-bell fizzy-ology or med-a-sin prize.

08.10.2025 04:18 👍 1 🔁 0 💬 0 📌 0
Preview
What if local control can actually help build housing? Some places want to build. Maybe we should just let them.

What if local control can actually help build housing?

27.09.2025 04:40 👍 8 🔁 3 💬 2 📌 2
Preview
Building AI for cyber defenders How we've improved Claude's cyber defense skills

Building AI for cyber defenders Oct 3, 2025
www.anthropic.com/research/bui...

05.10.2025 12:00 👍 0 🔁 1 💬 0 📌 0
Preview
Medicine’s AI Knowledge War Heats Up The Battlegrounds May Surprise You

My latest Substack: As OpenEvidence sprinted past UpToDate to become medicine's best AI-enabled point-of-care tool, I've been waiting for UTD's response. Now it's here, raising fascinating questions about which knowledge base should be our source of truth.
robertwachter.substack.com/p/medicines-...

01.10.2025 17:51 👍 7 🔁 2 💬 1 📌 0
Preview
How to make an antibody - Works in Progress Magazine Antibody therapies make up four of the world’s ten best selling drugs. If they were cheaper, they could prevent millions of deaths from rabies, malaria, and dengue.

Great piece by Alex Telford.

Antibody treatment is over a century old, when von Behring and Kitasato made an early 'vaccine' against diphtheria.

But back then, they came from injecting animals with toxins. Today they're produced by cells in steel tanks and treat millions. How did we get here?

18.09.2025 14:12 👍 34 🔁 11 💬 1 📌 0
Preview
Why science needs outsiders - Works in Progress Magazine Science has forgotten that the greatest breakthroughs often come from outsiders who are able to take a fresh perspective.

Outsiders are less bound by the norms & methods of a field. This gives them advantages: they are more likely to notice anomalies, they can bring ideas from other disciplines, and they can work for longer periods free from dominant intellectual trends worksinprogress.co/issue/why-sc...

19.09.2025 08:06 👍 4 🔁 1 💬 2 📌 0
Preview
How AI is shaking up the study of earthquakes “It's kind of like putting on glasses for the first time,” one scientist told me.

www.understandingai.org/p/how-ai-is-...

04.10.2025 16:29 👍 0 🔁 0 💬 0 📌 0
Post image

What will AI look like by 2030 if current trends hold?

Our new report zooms in on two things: (1) whether scaling continues (compute, data, power, capital), and (2) the capabilities this enables—especially for scientific R&D.

16.09.2025 16:51 👍 13 🔁 7 💬 1 📌 1
Preview
Real AI Agents and Real Work The race between human-centered work and infinite PowerPoints

AI agents are now capable of doing, real if limited work. But that work can be very valuable. For example, the new Claude Sonnet 4.5 was able to replicate published economics research from data files & a paper

We need to figure out what work is worth AI doing: www.oneusefulthing.org/p/real-ai-ag...

29.09.2025 18:56 👍 53 🔁 9 💬 2 📌 0

The time being evaluated by METR is not the time it takes the LLM to do the task, but how long it would take a human to do it.

30.09.2025 07:04 👍 1 🔁 0 💬 0 📌 0
Failing to Understand the Exponential, Again Posts and writings by Julian Schrittwieser

A good read about ai progress from an Anthropic researcher: www.julian.ac/blog/2025/09...

"It feels like the beginning of the covid-19 pandemic" in that the trends seem super clear once you're paying attention

29.09.2025 23:17 👍 13 🔁 1 💬 1 📌 0

You’re technically correct that by the time they named it COVID-19, it was no longer in the exponential growth phase. But SARS-CoV-2 did grow exponentially from late 2019 to late-Mar 2020, as reported cases grew to ~1M. Those are the early weeks he’s referring to.

30.09.2025 07:02 👍 2 🔁 0 💬 3 📌 0
Preview
How Common Is Accidental Invention? One of the most important inventions of the 19th century was mauve dye, the first synthetic aniline dye.

How Common Is Accidental Invention?

25.09.2025 12:01 👍 1 🔁 1 💬 0 📌 0
Preview
Measuring the performance of our models on real-world tasks We’re introducing GDPval, a new evaluation that measures model performance on economically valuable, real-world tasks across 44 occupations.

#OpenAI is introducing GDPval, a new way to measure the performance of models on real-world tasks.

It aims to help determine the broader impact of AI in society by focusing on economically valuable tasks.

#AI #LLM #GenAI #ChatGPT

25.09.2025 17:24 👍 1 🔁 1 💬 0 📌 0