te_chris

I’ve reimplemented some python science code into rust using LLMs and it absolutely works well if you setup the criteria for done. In my case it was GIS-ish code so a key requirement was identical in/out pairs at 10k points across CONUS. It worked, and when it didn’t it showed up bugs that the agent would go and research, read up on other implementations, then try again.

Net result faster (10-70x) and massively lower memory footprint. Now we’ve got a service that can do something essentially instantly that most people wouldn’t have even bothered to try before.

I think coding is done, as a trade practice. But a lot of jobs have sucked for a while anyway - endless api glue, js tedium.

There’s now a chance to do more with software.

I actually think art is safe. The machines don’t value it but we do. There’s something in that.

Yes, definitely load bearing.

Yes I agree. I got it to vibe up a simple react router app. When it crashed it was obvious that it had totally swallowed all errors in the name of a tidy error page. Getting it to re-add logs and debuggable errors was an exercise in patience as astra just got more and more tweaked while trying to solve the problem.

From an alignment perspective I’ve got no idea who it’s aligned to but it isn’t me, the meat proxy, who just wants to know why it crashed.

Public libraries pay for their books

Proust. In search of lost time (caveat I’m halfway through book one) is a story in which almost nothing happens but it is described exquisitely.

Because it’s a port of a scientific process and that’s what the scientists work in.

This, 100%. I’ve rewritten some hydrology code in rust using codex sol to first a) profile and create comprehensive tests of the python, b) create full benchmark suites, c) create full scientific benchmark suites, then d) port to rust using a few different techniques.

It works, it’s at least 5x faster, sometimes much more, and memory use is like 10x less and even less in cases where lots of map tiles are involved.

This shit rules.

What I’m wondering now is can we reliably evolve python and have codex act as an extremely unreliable transpiler to the rust.

Well, I can tell you this is making me actively reconsider dropping duckdb for ch as we go from prototype to prod

No, it’s a plague on anxious, over-optimised middle class parents.

The idea that someone can send over a schedule in advance is pure voodoo. Much like huckleberry, the ml app that makes new parents insane over wake windows.

This is a lovely piece of writing. Well done.

“What it is today”.

Oh, glad we managed to not miss out on the 2-sided ai slop panopticon

It’s great and the deep compiler integration means with a skill the agents write it well

I think there's a secondary competitor to certain types of process-SaaS like Vanta for smaller orgs: a smart Fractional (CISO, in the compliance case) paired with good AI. The purchaser gets proper human judgement, and you can handle the record-keeping bit with an ai-enabled DB like notion.

As a kiwi he was the best of us. Creative, talented, willing to roll up his sleeves, maker of exceptional wine. Haere ra

Spoilsport

I agree. Better to let it rip in a sandbox then spend your time correcting the finished product.

Waste of time being in the middle.

With pi they’re an extension, but that’s pi

The tweet Carmack's replying to is such a gross, cloying example of LLM slop. Bellard, of course is a legend.

GCP cloud run is pretty close to this. We’re using it now and I’ve a lot of experience with gke.

They’ve announced persistent “instances” recently which solves a big problem for us - sometimes you want continual long running workloads.

Great, so nihilism

Was going to say, open qwen in lm studio, say hi, watch the thinking traces

Nice. I've got Claude code teaching me maths at the moment. Some skills, file system learning log, oss text books and a full curriculum based on Math Academy (which I was doing before but got bored of) and UK high school and university. Teaching me concept-first though. So we start with something complex and go up and down until I get it all. It's not necessarily thorough for everything, but the depth of my intuition is much better and each time I use it I find myself unlocking another sector of the map. I love LLMs for this.

To nit pick: Micromobility is the opposite of this.

Extremely weird take.

Let me get this straight, they just didn’t supervise exams? For over 100 years. Lol.

I’ve never seen a billboard for it here - lived here 11 years.

1) the models do generalise so concepts translate 2) languages with more opinionated semantics and a better, more coherent community seem to be better. Python is a broad shitshow with multiple ways to achieve the same thing. Elixir is tight and focused. Claude is much better at elixir.

No, the personal injury lawyer billboards.

A thankfully American reference