My employer got snapped like a twig and we all suddenly lost our jobs. Anyone with opportunities or looking for connections up for a chat?

I'm a data person with experience working with on-prem, cloud, and air-gapped environments. Most recently I was embedded Data and MLOps/AI Engineer in a cracking little Data Science team, working with ongoing time series commercial banking transaction data covering several million customers.

LinkedIn profile is up-to-date enough to act as a CV: https://www.linkedin.com/in/jimgar/

If you had to know one thing about me: I take most pleasure at work in seeing my colleagues thrive, no matter which team they're in or level they are at.

UK only, Edinburgh or remote, but if it's hybrid 1/2 days a week I could commute to e.g. Glesga.

#GetFediHired

A poem about losing my job in the Smart Data Foundry insolvency.

You were doing good work?
Your colleagues weren't berks?
You built a nice lakehouse
But no one gets the perks?

A uni spokesperson croons
Such unfortunate news!
Our thoughts are with you
Anyway, toodle-do.

So it comes time again
To ask all my friends
Anything going?
Or shall I get in the bin...

CV is updated
And I've cuddled my cat.
Blog posts are coming
DuckDB, stuff like that

I’ve been AI this whole time. An Imbecile.

why, god? why is it

```shell
zip stuff.zip *.csv
```

and not

```shell
zip *.csv stuff.zip
```

I'm having "angry fun" imagining a world where we switch up the autistic and neurotypical populations. I want neurotypicals to see how they sometimes sound, by putting the shoe on the other foot:

Hi John. I hope it's ok to message. I heard from Sally that you have been diagnosed with Neurotypicalism Spectrum Disorder. I was surprised - you don't look Typicalist to me. It is worrying how many people are being diagnosed with Typicalism these days isn't it, but more power to you, I'm sure you got diagnosed properly.

Mastodon: Generally every day there's something interesting or uplifting on my timeline.

LinkedIn: Immediately get shown a blog post where someone says that "one good mid to senior engineer can now outpace a whole team and I think they should be one-person teams to move fast. Maybe everything gets broken down to one man teams." Stay away from me.

RE: https://hachyderm.io/@jimgar/117027943637566975

Please clap

If you want to know how viable using ChatGPT, Claude Opus/Sonnet, and (maybe) Gemini is right now, consider this. All three of those - with lots of different versions of ChatGPT in particular - are available for staff and students of The University of Edinburgh through the uni's Edinburgh Language Model (ELM) service. They also have a couple of locally hosted models, EuroLLM and Llama 3.3.

Recently they introduced Claude Opus. It didn't take long for a banner to appear at the top of the site. You could hear the hand wringing. It said something like, "Please reserve Opus for your most complex tasks, it is very expensive for the University. Prefer Sonnet if possible."

Weeks, maybe a couple of months go by, and today they've added Qwen 3.5 397B as another locally hosted option. The banner now reads (verbatim): "In the interests of reducing the university's spend on proprietary Generative AI suppliers please choose our new Qwen 3.5 model instead. Qwen (and our other self-hosted models) provide an inexpensive and environmentally more friendly option. **Please choose our Qwen model as default if you can.**"

I support the uni making more locally hosted models available. Leaving aside that the local models still come from specialist third parties, it's a good thing that they want to reduce reliance on the big 3 American suppliers.

It's a shame that the ultimate driver for that is money. But here we are. Supposedly the most transformative technology since smartphones (or the internet?), meant to be so useful you can outright replace workers with it, and yet the frontier models are now so expensive that one of the most lauded English-language universities in the world has to beg its own staff and students not to use them.

jUsT a tOoL

@grimalkina If we’re not production engines, what are we?

What might we need and benefit from that a robot or factory component doesn’t?

I'm re-reading the first half of The Psychology of Software Teams because we are discussing passages as a group during our weekly team CPD sessions, and I'm providing the materials.

This statement at the top of page 24 just leapt off the page at me: "Human beings simply are not production engines".

What a brilliant summation of much contained in the book so far. We could all do with meditating on that sentence. Especially the managers and executives amongst us.

cc @grimalkina

Programming languages

- Arse tats
- C piss piss
- Twig
- Runt
- GuavaShit

Re some discussion happening around this read-worthy blog post: https://xn--gckvb8fzb.com/i-regret-migrating-to-codeberg/

I'm seeing some people say outside of the blog post that Codeberg are suddenly taking ideological decisions when they previously hadn't.

Codeberg are a community-led non-profit. Their very existence is ideological.

Regarding the actual blog post, I only want to critique this: Linus Torvalds says LLMs are just a tool, and accepts their use to some degree in the Linux kernel development. So what? Linus has said a lot of things. He's been an absolute bastard to people at various points over the years. Do we also think if that behaviour is good enough for him, it should be good enough for a community-led non-profit?

It will never stop being funny to me that experienced software developers will waver like a child when writing even the simplest SQL statements, but they're perfectly happy to work with JavaScript and TerraForm (and whatever else is responsible for mobile apps being 1GB in size)

Are you even autistic if a simple “And will there be anything else for ya?” in a shop doesn’t make you buffer for 10 seconds in pure confusion every time

Pizzicato Five’s “Happy End of the World” is still an incredible album. It’ll be 30 years old in 2027!!! I used to listen to this while playing Pokemon Yellow on the GameBoy. Lord, take me back 🙏🏻

https://youtu.be/09Ie6pM6ZJk

Hey @hrbrmstr can I ask have you seen Zen (not the browser) before? What do you think? https://irbis.sh/zen

Not an editor command: W

SHOULD be an editor command: W

#Vim #QuestionableTypingSkills

Last night I gave a lightning talk about ROLLUP and CUBE at the @PostgresEDI Edinburgh PostgreSQL meetup. Code and slides are at https://github.com/jimgar/pg-rollups

If you have some groups and want aggregate stats for them, you probably reach for GROUP BY. But what you may **need** is ROLLUP or CUBE, both of which elegantly extend what's possible with grouped aggregates. If that piques your interest, the official docs on the topic are very good! https://www.postgresql.org/docs/current/queries-table-expressions.html

I'm giving up on programming related professional development in the evenings. I've started to learn Mandarin, instead.

Yes, this is driven by disillusionment with the industry, in large part thanks to AI. Professional development in evenings feels pointless. For 6 years coding has been my main special interest, and because I discovered programming in my 30s, I've also felt a pressure to catch up, learn, fill in gaps.

I like the process of writing code. I like the literal typing out of the words and symbols. I enjoy 'thinking by doing'. It was the same when I did lots of observational drawing as an animation student. There's a connection between mind and fingers that stimulates my brain like crazy.

That specific feeling of creating something is NOT the same as iterating with an agent or plain old LLM. I'm not saying using them makes you stupid, and I like what @grimalkina has been doing with her Claude skills in that respect (you can listen to her [podcast ep](https://www.changetechnically.fyi/2396236/episodes/18692591-you-can-learn-with-ai) about it), I'm just saying they're qualitatively different experiences for me.

I don't enjoy coding with agents very much. It's kind of hollow. I'm still in charge and have to do most of the planning, yes, so I'm still thinking about stuff. But again, it's not the same kind of thinking. It's missing the part that relies on the doing.

The other thing is, and this may get me crucified because I'm saying it aloud on the fediverse, but, models are way, way, way better than they were even 18 months ago. They were pathetically bad. A waste of time. Now, if you work in Python and all you need is something a mid-level developer could understand and work with, an agent in your IDE using one of the latest models can probably do it (with lots of guidance).

I also think it's also reasonable that the outcomes of using them will get even better, if for no other reason than clever people will improve how we interact with them and add extra tooling about important things, like correctness of code outputs.

I'm a very valuable programmer to have in a team, but I'll only ever be mid at the actual programming. So all told, the magic is gone and I no longer feel an impetus to keep up.

Maybe one day I can be mid at speaking Mandarin, as well as programming.