Ask ChatGPT how to build a bomb, and it will flatly respond that it “can’t help with that.” But users have long played a cat-and-mouse game to try to trick language models into providing forbidden information. These “jailbreaks” have run from the mundane — in the early years, one could simply tell a model to ignore its safety instructions — to elaborate multi-prompt roleplay scenarios. Source Ask ChatGPT how to build a bomb, and it will flatly respond that it “can’t help wit...

Probably not a popular thing to say today. The zeitgeisty thing to say is that we should all log off and live terrible cottagecore solarpunk lives raising chickens and being mindful. I wish people were more online and more public. I have rarely wished the opposite. Consider this post addressed to you, the reader. Your Writing I will often find a blog post on Hacker News that really resonates. And when I go to check the rest of the site there’s three other posts. And I think: I wish you...

This document curates the most common questions Shreya and I received while teaching 700+ engineers & PMs AI Evals. Warning: These are sharp opinions about what works in most cases. They are not universal truths. Use your judgment. 👉 Want to learn more about AI Evals? Check out our AI Evals course . It’s a live cohort with hands on exercises and office hours. Here is a 25% discount code for readers. 👈 Listen to the audio version of this FAQ If you prefer to listen ...

How does a distributed system reliably determine when one of its members has failed? This is a tricky problem: you need to deal with unreliably networks, the fact that nodes can crash at arbitrary times, and you need to do so in a way that can scale to thousands of noes. This is the role of a failure detection system, and is one of the most foundational parts of many distributed systems. There are many rather simple ways to solve this problem, but one of the most elegant solutions to distrib...

Elon Musk is not happy with the EU fining his X platform and is currently on a tweet rampage complaining about it. Among other things, he wants the whole EU to be abolished. He sadly is hardly the first wealthy American to share their opinions on European politics lately. I’m not a fan of this outside attention but I believe it’s noteworthy and something to pay attention to. In particular because the idea of destroying and ripping apart the EU is not just popular in the US; it’s popu...

Early microprocessors were very slow when operating with floating-point numbers. But in 1980, Intel introduced the 8087 floating-point coprocessor, performing floating-point operations up to 100 times faster. This was a huge benefit for IBM PC applications such as AutoCAD, spreadsheets, and flight simulators. The 8087 was so effective that today's computers still use a floating-point system based on the 8087. 1 The 8087 was an extremely complex chip for its time, containing somewhere betwee...

The DC-ROMA II is the fastest RISC-V laptop and is odd Inside this Framework 13 laptop is a special mainboard developed by DeepComputing in collaboration with Framework. It has an 8-core RISC-V processor, the ESWIN 7702X—not your typical AMD, Intel, or even Arm SoC. The full laptop version I tested costs $1119 and gets you about the performance of a Raspberry Pi. A Pi 4—the one that came out in 2019. Jeff Geerling December 8, 2025 The DC-ROMA II is the...

A vulnerability in a docker container caused my VPS to catch a bug A vulnerability in a docker container caused my VPS to catch a bug

Whatever it is, let me start it with gratitude. Gratitude is fertile ground. Put in the seeds of your dreams and desires. Keep the ground watered and pull the weeds. Soon the seeds will grow. (Conversely, worry is fertile ground for all your fears. Stay worried and you will harvest an abundance of fears.) Gratitude has nothing to do with what you have, how good or easy you’ve got it, whether you get what you want or don’t. Gratitude is not concerned with such petty measurements of valu...

I talked to Brendan Samek about Canada Spends , a project from Build Canada that makes Canadian government financial data accessible and explorable using a combination of Datasette, a neat custom frontend, Ruby ingestion scripts, sqlite-utils and pieces of LLM-powered PDF extraction. Here's the video on YouTube . Sections within that video: 02:57 Data sources and the PDF problem 05:51 Crowdsourcing financial data across Canada 07:27 Datasette demo: Search and facets ...

When people think about autonomy in aviation, they often focus on what it can do — flying complex missions, reacting to dynamic environments, making intelligent decisions in real time. But the real challenge isn’t getting autonomy to work , rather it’s proving that it can be trusted.   For autonomous flight to scale across aircraft and mission sets, it must meet the same airworthiness expectations we apply to human pilots. This isn’t just paperwork; it’s the difference between a de...

There's an old compiler-building tutorial that has become part of the field's lore: the Let's Build a Compiler series by Jack Crenshaw (published between 1988 and 1995). I ran into it in 2003 and was very impressed, but it's now 2025 and this tutorial is still being mentioned quite often in Hacker News threads . Why is that? Why does a tutorial from 35 years ago, built in Pascal and emitting Motorola 68000 assembly - technologies that are virtually unknown for the new generation of progra...

In 2023 Robb Knight started the defaults trend, covering what people used as their default apps for specific purposes. I did a 2023 version, and while I mentioned planning to do a late 2024 version in my 2024 year end wrap up post, I never wound up getting to it. I’ve got a 2025 edition of my default apps below, with changes in bold. PC Tablet Phone Mail Client Proton Web N/A Proton App Mail Server ^ ^ ^ Notes Joplin Joplin Joplin Tasks N/A N/A Tasks (org. In 2023 Robb Knight started the defa...

Do Not Optimize Away Dec 9, 2025 Compilers are sneaky beasts. If you time code like this: var total: u32 = 0 ; for ( 0 ..N) | i | total += i; print( "total={}" , .{total}); You will discover that LLVM is as smart as a little kid named Gauss, and replaces the summation with an equivalent formula N ( N + 1 ) 2 . What’s more, if you write something more complicated like total += i + 2*i*i - i*i*i , you’ll see that LLVM figures out...

Last week’s now-corrected headline & intro of a peak-peek at lunar samples had a peak typo (pun intended). I wish I could conveniently blame it on the very productive yet equally tiring Hong Kong trip to cover the international lunar sample science symposium coupled with the excitement of having seen fresh Moon samples. Or perhaps put it on Hong Kong itself because you can’t peek at its towering structures—they peak at you. But the reality is that it was just me being sloppy while rewr...

The playwright  Tom Stoppard passed away at the age of 88 on Nov. 29, 2025. ONE) He wrote many plays and some movies.  Below I highlight his works whose themes I think will be of interest to my readers (Or at least to me—your mileage may vary.) 1)  Rosencrantz and Guildenstern are Dead (1966) This is Hamlet told from the point of view of two minor characters who, in Shakespeare’s original, can best be described as plot furniture. The play begins with, R and G are flipping coins. R bets ...

Following my posts on determinism and durable function trees , this installment advances this blog post series “The Theory of Durable Execution”. Durable execution engines (DEEs) talk about “workflows”, “activities”, “virtual objects”, “handlers”, and “functions”, but they’re often describing the same underlying execution patterns. This post proposes a model that extends the generic durable function into three forms: stateless functions, sessions , and actors ....

One of my American colleagues said this in the context of a new client: We can propose a few solutions, but let’s not flounder in the Lake of Options. This is awesomesauce. We can propose a whole bunch of options, but each of these require specs, cycling, feedback, iterations, docs, consultation, and other processes. People hire experts to get their expert opinion, and selecting the best of many options for their requirements. Too many options burn time, effort, and cashmoney. I can ev...

For a while, I have had an idea for Artemis – the calm web reader I maintain – to have a page that shows your feed and feed settings on the same page. I am interested in this idea because there is presently a distance between the settings to customise your reader and the reader itself. You need to click back-and-forth to see the impact of a change in settings to your feed. As a user, I would prefer to be able to immediately preview the impact of changing my feed and reader interface, befor...

Are Two Heads Better Than One? You're playing a game with your lying friends Alice and Bob. Bob flips a coin and shows it to Alice. Alice tells you what she saw - but she lies 20% of the time. Then you take your best guess on whether the coin is heads or tails. Your best strategy is to trust whatever Alice says. You're right 80% of the time. Now Bob joins in. He makes up his mind independent of Alice, and he _also_ lies 20% of the time. You were right 80% of the time by trusting Alice. How m...

One of the fundamental things that query planners have to do is decompose predicates . What do I mean by that, well, it's actually something that comes up quite often in everyday programming. Imagine we have a database and a relation with two columns, a and b , along with two indexes: one on a , and one on b . We can write some code to simulate this. First, we can create a dataset of a s and b s: let data : Vec = ( 0 .. 1000000 ) . map ( | _ | ( random_range ( 0 ....

In the past two months or so, I spent some time making tiny GLSL demos. I wrote an article about the first one, Red Alp . There, I went into details about the whole process, so I recommend to check it out first if you're not familiar with the field. We will look at 4 demos: Moonlight , Entrance 3 , Archipelago , and Cutie . But this time, for each demo, we're going to cover one or two things I learned from it. It won't be a deep dive into every aspect because it would be extremely redun...

There are only two hard things in computer science: cache invalidation and naming things. — Phil Karlton With respect, I think that handling interrupts is harder than either of these. Yesterday’s post explained how SimPy does this. Today, after several near misses, we’ll look at how to add it to our simulation. A Quick Recap Our Simulation class now includes a process that waits a random interval, chooses a random developer, and interrupts her by calling .interrupt : cla...