dgunay

6 ms·

1.
by dgunay 8d ago
Haven't touched it in a few years but Bevy was a really nice breath of fresh air when I dipped my toes back into hobby game dev. I wasn't used to being able to create a game almost purely as code without having to touch a WYSIWYG
2.
by dgunay 8d ago
For things of global relevance, I keep a git repo in ~/.skills. It has a script that syncs skills across agent harnesses when run. If I think of an interesting skill that seems like it'd be broadly useful, I put it in this repo. I
3.
by dgunay 13d ago
This is a big reason why I feel like even though LLMs are _effectively_ AGI in some regard, they also are a hack around what most people figured AGI would look like before the advent of LLMs. Humans can do metacognition, output multimodally
4.
by dgunay 20d ago
Luna max is suitable for like 90% of the kinds of code changes I want to make. I only find myself actually reaching for a Sol or Fable tier model if the problem is very complex. If you're willing to build the guardrails and do some ext
5.
by dgunay 20d ago
Everyone has their own distinct mix of jargon and uses some words more or less than others, but now it's as if you have 1000 coworkers who are all the same person, and talk the exact same way, and it just so happens that their writing
6.
by dgunay 21d ago
Oh definitely, there's great value in it for those who already own Macbooks for one reason or another. I just personally have found myself shying away from buying a Macbook multiple times over the years because the Asahi Linux experien
7.
by dgunay 21d ago
I have immense respect for the incredible work that the Asahi team does, but at the rate that Intel and AMD have been catching up on power efficiency I wonder whether anyone will really _have_ to run Linux on M series laptops. There was a c
8.
by dgunay 28d ago
You can do pipelined option/result manipulation now, but without sum types/pattern matching it's not going to be quite as useful as Rust's.
9.
by dgunay 1mo ago
Oh no the models are absolutely getting better, I'm just amazed that only 6 months ago I was using gpt-5.3-codex, and now I can use gpt-5.6-luna for similar results at like 1/15th the cost. Now 5.6-sol is being slashed by 50%? Ama
10.
by dgunay 1mo ago
I'm loving this race to the bottom.
11.
by dgunay 1mo ago
I like Go but a couple of these "advantages" wash out when you add the scale and typical usage patterns of agents. | Go is Readable / Go is Maintainable It's true that Go, as a low-magic language, tends to be very same-y
12.
by dgunay 1mo ago
Times I have needed to do this: - An agent messed with my git config for some reason so I had to go back and fix the commit authorship - I created a PR mostly based on work that someone else did, so I reauthored most of the commits to
13.
by dgunay 1mo ago
Besides the debate about whether this is "safer" than manual human review, I have a slightly different problem. Very often, when I'm running Claude in manual review mode, it will attempt to do things which are not "dange
14.
by dgunay 1mo ago
It's less the tmux config and more the scripting around it. (shameless plug: I launch agents in tmux directly into docker containers with remuda, a tool I built at work & open sourced [0]) Anyway, I have a vibe coded script that sh
15.
by dgunay 1mo ago
Over the last year I ended up duct taping a lot of functionality onto tmux to get what these new age multiplexers set out to do (session launch, live agent status, etc). And it works, but it definitely feels like duct tape. I'm not sur
16.
by dgunay 1mo ago
In a lot of cases there is, but you have to be aggressive about allowlisting commands. It can also be difficult to predict when being able to do a read-only command goes from safe to part of a vulnerability chain. Also the permissioning sys
17.
by dgunay 1mo ago
For me the problems with agent permission prompts are twofold: 1) I generally have a lot of things where I am okay with the agent calling a specific tool (maybe in certain ways) as much as it wants. This allowlisting approach is often defea
18.
by dgunay 1mo ago
No, but if I got paid overtime I'd absolutely do it (or something more reasonable like 50-60 hrs/week)
19.
by dgunay 1mo ago
It sounds too much like "parasite" for me to be comfortable using it. I'm assuming some random person made it up. It's okay to just say "people of/from <place>", guys.
20.
by dgunay 2mo ago
At least where I'm at, you can still go cheaper for pizza and burgers. In-N-Out's combo meals start at about $8. If I want to, I can also get 2 big slices of serviceable pizza for a bit less than $10 in a mom & pop shop, and t
21.
by dgunay 2mo ago
It's cool that they're doing this but with iterators and soon in 1.27 generic method parameters, you can already DIY most of this with little effort. I've got iterators at work for nearly everything I could want, small librar
22.
by dgunay 2mo ago
I have been seeing people use it more who I am fairly confident are not otherwise using AI to write public-facing communications. It's almost like there's a minor movement to try and reclaim it. Either that or it's somehow ge
23.
by dgunay 2mo ago
There's definitely a lot of AI-generated websites, especially for dev tools & SaaS, that love to have dark backgrounds, cool colors (purple especially is making a resurgence it seems), and tons of fancy UI widgets.
24.
by dgunay 2mo ago
I don't understand the challenges involved super well but I feel like this might be better as a piece of actual code (like a linter) that you can run to have an LLM iteratively de-slop its own writing, rather than forcing it out of dis
25.
by dgunay 2mo ago
Maybe this is naive or startup-brained of me: I understand not wanting HR-related things to be potentially exposed by a misaligned agent, but when it comes purely to the building of software I personally do not care at all who can read the
26.
by dgunay 2mo ago
I haven't tried fable in this role yet but overall I've found most models to have a preferred way of doing orchestration tasks and in general do not consistently follow instructions that tell them to work in a certain way. Especia
27.
by dgunay 2mo ago
I'm guilty of this too, but at least my speech tics are mostly a unique blend to me, and they also tend to change seasonally. Meanwhile the emdashpocalypse has been going on for years.
28.
by dgunay 2mo ago
DMS is a pretty good gateway drug out of major DEs and into WMs like Niri. I had already started using Niri but the base DMS configuration has a lot more batteries included so I switched and so far haven't had any major complaints.
29.
by dgunay 3mo ago
That's definitely valid, for b2c applications though a lot of the time you will never deal with numbers of that magnitude.
30.
by dgunay 3mo ago
All of the data we control (in the database, in our apis, etc) is integer cents. When we have to interface with a system that represents money using JSON numbers as dollars.cents, we parse or serialize it into an arbitrary precision decimal

More ›