I don’t know rust, but for example in Swift the type system can make things way more difficult.

Before they added macros if you wanted to write ORM code on a SQL database it was brutal, and if you need to go into raw buffers it’s generally easier to just write C/objc code and a bridging header. The type system can make it harder to reason about performance too because you lose some visibility in what actually gets compiled.

The Swift type system has improved, but I’ve spent a lot of time fighting with it. I just try to avoid generics and type erasure now.

I’ve had similar experiences with Java and Scala.

That’s what I mean about it being nice to drop out of setting up some type hierarchy and interfaces and just working with a raw buffers or function pointers.

“These aren’t real candidates. They aren’t campaigning. They aren’t engaging with constituents,” Poilievre wrote.

Boy, that sounds a whole lot like how none of the CPC candidates showed up to any of my local debates or showed up on a single local news or radio program, or took a single interview.

Frankly, I don’t think we should be limited from running for office based on their percieved level of seriousness or that it’s a protest. That’s a slippery slope right there. We can increase the signatures or whatever, but it’s still not going to be particularly useful and will primarily increase the floor of how much money you need to actually campaign.

I actually do like that C/C++ let you do this stuff.

Sometimes it’s nice to acknowledge that I’m writing software for a computer and it’s all just bytes. Sometimes I don’t really want to wrestle with the ivory tower of abstract type theory mixed with vague compiler errors, I just want to allocate a block of memory and apply a minimal set rules on top.

Reminds me of java

I have Toolkit toolkit = Toolkit.getDefaultToolkit(); seared into my brain. Then there were the bean factories…

Now do Epstein!

It’s a bit passed Canada Day but I like to see this

For over 100 years we were the best of allies

It took weeks for him to throw all of that away.

You don’t declare “economic warfare” on your friends. You don’t threaten to annex them, tell them they’re not a viable country, tell them they’re destined to fall, and still expect friendship.

And it’s not just Trump to blame. The voters and the entire US government have allowed that behaviour and enabled him.

If they’re getting upset about our boycotts that just means boycott even harder. They said they don’t need us, now they can see what that’s like.

I tried following my cat and all he lead me to was his food bowl.

Oh wow, I didn’t know you were having these problems with him.

But since you’re in my office let’s start you on a PIP and I have to write you up for this hostility against a fellow employee.

Batch process turning unstructured free form text data into structured outputs.

As a crappy example imagine if you wanted to download metadata about your albums but they’re all labelled “Various Artists”. You can use an LLM call to read the album description and fix the track artists for the tracks, now you can properly organize your collection.

I’m using the same idea, different domain and a complex set of inputs.

It can be much more cost effective than manually spending days tagging data and writing custom importers.

You can definitely go lighter than LLMs. You can use gensim to do category matching, you can use sentence transformers and nearest neighbours (this is basically what Semantle does), but LLM performed the best on more complex document input.

One built a bunch of local search tools with MCP and that’s where I get a lot of my value out of it

RAG workflows are incredibly useful and with modern agents and tool calls work very well.

They kind of went out of style but it’s a perfect use case.

The tool isn’t returning all code, but it is sending code.

I had discussions with my CTO and security team before integrating Claude code.

I have to use Gemini in one specific workflow and Gemini had a lot of landlines for how they use your data. Anthropic was easier to understand.

Anthropic also has some guidance for running Claude Code in a container with firewall and your specified dev tools, it works but that’s not my area of expertise.

The container doesn’t solve all the issues like using remote servers, but it does let you restrict what files and network requests Claude can access (so e.g. Claude can’t read your env vars or ssh key files).

I do try local LLMs but they’re not there yet on my machine for most use cases. Gemma 3n is decent if you need small model performance and tool calls, phi4 works but isn’t thinking (the thinking variants are awful), and I’m exploring dream coder and diffusion models. R1 is still one of the best local models but frequently overthinks, even the new release. Context window is the largest limiting factor I find locally.

Vibe coding you do end up spending a lot of time waiting for prompts, so I get the results of that study.

I fall pretty deep in the power user category for LLMs, so I don’t really feel that the study applies well to me, but also I acknowledge I can be biased there.

I have custom proprietary MCPs for semantic search over my code bases that lets AI do repeated graph searches on my code (imagine combining language server, ctags, networkx, and grep+fuzzy search). That is way faster than iteratively grepping and code scanning manually with a low chance of LLM errors. By the time I open GitHub code search or run ripgrep Claude has used already prioritized and listed my modules to investigate.

That tool alone with an LLM can save me half a day of research and debugging on complex tickets, which pays for an AI subscription alone. I have other internal tools to accelerate work too.

I use it to organize my JIRA tickets and plan my daily goals. I actually get Claude to do a lot of triage for me before I even start a task, which cuts the investigation phase to a few minutes on small tasks.

I use it to review all my PRs before I ask a human to look, it catches a lot of small things and can correct them, then the PR avoids the bike shedding nitpicks some reviewers love. Claude can do this, Copilot will only ever point out nitpicks, so the model makes a huge difference here. But regardless, 1 fewer review request cycle helps keep things moving.

It’s a huge boon to debugging — much faster than searching errors manually. Especially helpful on the types of errors you have to rabbit hole GitHub issue content chains to solve.

It’s very fast to get projects to MVP while following common structure/idioms, and can help write unit tests quickly for me. After the MVP stage it sucks and I go back to manually coding.

I use it to generate code snippets where documentation sucks. If you look at the ibis library in Python for example the docs are Byzantine and poorly organized. LLMs are better at finding the relevant docs than I am there. I mostly use LLM search instead of manual for doc search now.

I have a lot of custom scripts and calculators and apps that I made with it which keep me more focused on my actual work and accelerate things.

I regularly have the LLM help me write bash or python or jq scripts when I need to audit codebases for large refactors. That’s low maintenance one off work that can be easily verified but complex to write. I never remember the syntax for bash and jq even after using them for years.

I guess the short version is I tend to build tools for the AI, then let the LLM use those tools to improve and accelerate my workflows. That returns a lot of time back to me.

I do try vibe coding but end up in the same time sink traps as the study found. If the LLM is ever wrong, you save time forking the chat than trying to realign it, but it’s still likely to be slower. Repeat chats result in the same pitfalls for complex issues and bugs, so you have to abandon that state quickly.

Vibe coding small revisions can still be a bit faster and it’s great at helping me with documentation.

Just make your JIRA tickets increasingly more detailed.

You get to procrastinate but it looks like you’re doing stuff

It’s incredible that it works, it’s incredible what just encoding language can do, but it is not a rational thinking system.

I don’t think most people care about the proverbial man behind the curtain, it talks like a human so it must be smart like a human.

They could hire on a contractor and eschew all those costs.

I’ve done contract work before, this seems a good fit (defined problem plus budget, unknown timeline, clear requirements)

I explicitly told it eleven times in ALL CAPS not to do this. I am a little worried about safety now.

Well then, that settles it, this should never have happened.

I don’t think putting complex technical info in front of non technical people like this is a good idea. When it comes to LLMs, they cannot do any work that you yourself do not understand.

That goes for math, coding, health advice, etc.

If you don’t understand then you don’t know what they’re doing wrong. They’re helpful tools but only in this context.

What is the limit of transformative works here?

If I build an auto encoder neural network and train it to learn and return content exactly, that’s probably not allowed, but if I train it to rephrase the news, is that allowed?

I don’t want to alarm you, but do not approach the egg lest a tiny face hugger will jump out at you