machine_learning

Coding agents and deep learning

Andy sent me this post about how Carl Kolon uses and thinks about coding agents. The post is pretty interesting, and when I went to reply to Andy I realized that at seven paragraphs perhaps the reply is better suited to here than a signal message. Also, because that is more friendly than replying to messages with links to your blog to drive traffic and the adoration of your readers? This is of course a forward looking statement for me, I do indeed hope one day to have a reader here but baby steps. Similarly to Carl I certainly started using LLMs as "smarter search", cutting and pasting queries into claude.ai, and waiting patiently for an answer. I think this should make Google very worried, especially as its so good at finding answers. The improved performance over a "raw Google" is largely because of persistence -- the LLM doesn't perform  single search, it will keep searching until it finds the answer. The decreasing levels of supervision Carl talks about is what I am talking about when I talk about prompts and planning, which has been my favourite LLM topic for the last few weeks. Increasingly I am writing a plan for…

Continue ReadingCoding agents and deep learning

I think I just experienced my first hallucinated security vulnerability

Yesterday my time I experienced what I think was my first LLM hallucinating a responsible security disclosure. Honestly it was no curl situation, but I think it was still interesting. The bug is on launchpad.net if you're interested in taking a look. I think in total I spent a couple of hours on the whole thing, with the hardest bit being trying to understand what the author was claiming. Fundamentally they had conflated being able to change the state of memory and other hardware inside their virtual machine with changing the state of those things for the hypervisor. They did not seem to understand that the video memory of the guest was not the video memory of the host for example. That said, I tried to be nice and I hope my replies were perhaps a little useful to them.

Continue ReadingI think I just experienced my first hallucinated security vulnerability

Ancient code, mental health, and AI tooling

In the early 2000s I was in my mid 20s, working a dead end job as a Windows programmer, and had two very young kids who were not super good at sleeping. I had worked as what I would now call a systems programmer for the Australian patents and trade marks office for a few years in the late 1990s doing low level image manipulation code -- we had a for the time quite impressive database of scanned images of patents and trademarks, and sometimes we need to do things like turn them into PDFs or import a weird made up image format from the Japanese patents office. So when you combined those things -- previous experience in a field I found interesting, a job I did not currently find interesting, and a lot of spare time very early in the morning because the kids wouldn't sleep but my wife really did need a rest -- you end up with a Michael who spent a lot of time writing image manipulation code on his own time. Even back then I was pretty into Open Source, so I released what I think was probably the first Open Source PDF generation library,…

Continue ReadingAncient code, mental health, and AI tooling

Talking to people about AI is hard

I am not entirely sure what makes the AI debate so polarizing, although I suspect it has something to do with feeling threatened by a changing landscape. What I can say with certainty is that I find having a nuanced conversation about AI with people often difficult. It seems to me that people fall into two polar opposite camps — those who thing AI is completely great and that we should fire all the programmers and creatives; and those who think that AI is all bad and we should go backwards in time to a place before it existed.

(more…)

Continue ReadingTalking to people about AI is hard

A self drive computer science degree, 2025 edition

What is you want to study computer science, but can’t afford university fees? Or, studied computer science a really long time ago and want to see what’s changed? Or just like learning stuff? Well, lots of schools now post their lectures on YouTube, so its entirely possible to construct a zero cost “self driving degree”, as long as you’re good enough at Canva to make your own certificate at the end. I consider this list incomplete, but in the end I decided I’d post the things from 2025 that I’d found and liked. I can always do an updated version later.

(more…)

Continue ReadingA self drive computer science degree, 2025 edition

Writing my own KVM client in python

I was talking to a friend the other day about our shared mutual appreciation of virtio-vsock, and it made me wonder something. How do virtual machines on Linux actually work? I know it involves qemu and the kernel’s KVM virtual machine implementation, but exactly how do they interact? How does the kernel get qemu to do emulation tasks as required?

qemu is several things hanging out together in a trench coat, but one of those things is software which can configure Linux’s built-in KVM virtual machine functionality to run a virtual machine, and then handle emulation of the devices that virtual machine is attached to which cannot be represented with actual physical hardware. This part of qemu is called a “KVM client” in the Linux kernel documentation. Its called that because if we ignore the emulation part for now it is just literally a client calling established APIs to the Linux kernel.

(more…)

Continue ReadingWriting my own KVM client in python

Tensorflow, ROCm, and version mismatches

I’ve fallen into this pattern where I do an hour or so of self-directed learning in the mornings before going to work. Until recently it was an excellent CMU course on the design of SQL database systems, which I’ve mentioned previously here. I’ve finished that, so I thought I would do something shorter and fun as a break before finding another course to do. I chose The freeCodeCamp.org hot dog or not hot dog tensorflow course. 90 minutes seemed achievable, and I too wish to know if an object in front of me is a hot dog or not.

(more…)

Continue ReadingTensorflow, ROCm, and version mismatches

High level themes from CloudCon 2025

  • Post author:
  • Post category:Conference

I've just been in Sydney for a couple of days for CloudCon 2025. I think depending on how you count this is my third one of these events -- the event has changed names at least twice, so its actually a little hard to work out the lineage of the event. This year's conference was noticeably smaller than last years which is confusing to me for an event which is so competitively priced and branded so heavily with the hot topics dejour. That said the event was well run, in a good venue, and well worth the time. That is, this event really deserves more support than its getting. There were some clear themes from the event for me: ClickHouse is cool. Or at least I think so. ClickHouse observability certainly has potential, but I think the underlying SQL database is actually the most interesting bit. ClickHouse is also investing heavily in the Australian market right now, so I suspect they're seeing strong traction here. No one talks about "devops" any more, because its become a meaningless term where everything is devops if you squint at it right. Instead people are using the term "platform engineering", which doesn't appear to have…

Continue ReadingHigh level themes from CloudCon 2025

Taming Silicon Valley

  • Post author:
  • Post category:Book

The similarities and contrast between this book and AI Snake Oil are striking. For example, AI Snake Oil describes generative AI as something which largely works but is sometimes wrong, whereas this book is very concerned about how they’ve been rushed out the door in the wake of the unexpected popularity of ChatGPT despite clear issues with hallucinations and unacceptable content generation.

Yet the books agree on many things too — the widespread use of creators’ content without permission, weaponization of generative AI political misinformation, the dangers of deep fakes, and the lack of any form of factual verification (or understanding of the world at all) in the statistical approaches used to generate the content. Big tech has no answer for these “negative externalities” that they are enabling and would really rather we all pretend they’re not a thing. This book pushes much harder on the issue of how unregulated big tech is, and how it is repeatedly allowed to cause harm to society in returns for profits. It will be interesting to see if any regulation with teeth is created in this space.

(more…)

Taming Silicon Valley Book Cover Taming Silicon Valley
Gary F. Marcus
Computers
MIT Press
September 17, 2024
247

How Big Tech is taking advantage of us, how AI is making it worse, and how we can create a thriving, AI-positive world. On balance, will AI help humanity or harm it?

Continue ReadingTaming Silicon Valley

AI Snake Oil

  • Post author:
  • Post category:Book

Nick recommended I read this book, so here it is.

The book starts by providing an analogy for how we talk about AI — imagine that all transport vehicles were grouped by one generic term instead of a variety like “car”, “bus”, “rocket”, and “boat”. Imagine the confusion a conversation would experience if I was talking about boats and you were talking about rockets. This is one of the issues right now with discussions of “AI” — there are several kinds of AI, but the commentary is all grouped together and conflating the various types. I think this is probably a specific example of what Ben Goldacre talks about in Bad Science — science reporting by non-scientists is often overly credulous, and misses the subtleties.

(more…)

AI Snake Oil
Arvind Narayanan, Sayash Kapoor
2024
348

From two of TIME's 100 Most Influential People in AI, what you need to know about AI and how to defend yourself against bogus AI claims and products. Confused about AI and worried about what it means for your future and the future of the world? You're not alone.

Continue ReadingAI Snake Oil

End of content

No more pages to load

Older Posts