I’m conflicted, because hackathons are all about quickly proving if ideas works or not… which is exactly what the good coding agents excel at. We’re even always saying “never ever ship hackathon code, it’s slop meant to be deleted” which again, fits the exact mindset many (rightfully, usually) attribute to coding agent results.
It’s of course entirely unfair pitting agent-equipped teams against braincoders (idk what do you even call that?). If the coding agent team is at all competent with the tools, they’re likely gonna destroy braincoders on raw idea exploration speed.
Maybe I’m missing something, but I’m not at all surprised a coding-agent derived project won this? The Kaggle homepage leads strong with:
> Discover what actually works in AI. Join 32M+ builders, researchers, and labs evaluating agents, models, and frontier technology through crowdsourced benchmarks, competitions, and hackathons.
That is pretty straightforward about what happened here?
pretty much every day I feel that tickling of “what the hell did I just do?”
but I don’t care. I put 15 years into my tech career. I am good at building software, and I will not let this ridiculous “resume gap” problem stop me from taking a break for my mental health. any tech employer that wouldn’t want me because of it is a place I wouldn’t want to work anyway.
also, to be honest, I’m writing more code now than I ever did in the last year of my tech job… working on a full CMS and custom website combo for my friends bar, such that I can copy that template over for future projects (want to help local businesses escape the bullshit machine). also building a cool web development desktop app. and more! I’m having a great time
yeah, at a local pizza/taproom place. they share tips with cooks, works out to ~$35/hr. I'm not expecting it to replace a tech salary, but it helps offset the savings burn (which was specifically set aside for this). engaging with real people in real life about real things is also an indreeeedibly nice change of pace (even the stressful/tense moments of shit hitting the fan).
yeah! it's pretty cool when I start doing something that used to be difficult and now it's just... not. weird how long your old form sticks with you in your head like that.
realized this one about myself earlier this year, it really helped to boil it down to something besides “I am just inherently bad at things.” that attitude worked as a dumb single kid, but it was harming my adult life and relationships.
therapy helped me get there. I have been on bupropion for about a year, and recently started on methylphenidate. it might not be the right one for me, or maybe too small of a dosage. I’m taking it slow and being deliberate with the drugs.
working with a personal trainer to get in shape and lose weight, as well as quitting my fully vested tech job to fuck off and be a cook for a while has also done wonders for me here. it’s cliche, but you really can’t replace fitness with anything else, and that took me 35 years to internalize.
I have always wanted a web browser with tmux/iterm2 style tabs/panes, as well as the ability to save the window configuration and tab contents to a file that can be checked in to projects and loaded.
Load all of your project's documentation links, local development browser tools (database viewers, etc.) into a set of views that can be source controlled with the project. Don't force people to use their daily driver browsers for this, or hack side-by-side views together with their OS window managers. Zen and friends have split panes, but it's not the robust tab/panel system that I wanted.
There is a simple tab widget system, which so far has:
- Viewport manager: basically 1:1 with what is in Chrome devtools
- Session manager: create and manage browser sessions as a first-class entity, and attach/detach tabs to these sessions. Includes a simple "incognito" toggle as well.
- System light/dark override: stop flipping this on/off for your whole OS to test "system" light/dark mode (tedious)
- Reload trigger: pick a target tab. if that tab reloads, so does this one.
- Log file viewer: if the tab source is a local file, change the tab's view to a structured log file parser with search/filter, play/pause, etc.
- Screenshot/Video capture: not built yet. pretty self explanatory.
Great keyboard controls are a hard requirement for me. It's a little tricky since content in web views can capture this too, so I have a global "nav mode toggle" you enter to move around between panes and the tabs within them. Actively figuring out the correct UX, but I am liking what I have so far.
Toying with the idea of a "tab link" which allows you to store a set of "source" tabs in a view, but create "links" in other views, where the navigation is synced across all instances. Useful if you want to have, say, the Tailwind docs open to a specific page, but have that page shared across different views. For example, if you want to have one view specifically for mobile view work and another for desktop view work, and not have to manually navigate to the same Tailwind docs page in both views.
I'm honestly just using it as I work on another real project, and adding features as I think "hey wouldn't this be nice?" Which is a pretty fun and satisfying process. I don't have it published yet, because I'm not entirely sure if it's worth sharing at the moment, but I feel like I'll discover that along the way here and go from there. Maybe someone here will chime in :P
I just recently decided to replace iterm2 with wezterm when I started moving macbook over to nix. iterm2 is about the only one that didn’t work well for this, since you can’t source control the configuration (import/export doesn’t cut it)
Any of the ones you mentioned would probably work good with nix too. I don’t really care about the config being scriptable at all, it was just the first terminal that easily let me set all of the keyboard shortcuts I wanted, so I stuck with it.
Completely agree, I hate the “hackathon” for so many reasons, guess I’ll vent here too. All of this from the perspective of one frustrated software engineer in web tech.
First of all, if you want innovation, why are you forcing it into a single week? You very likely have smart people with very good ideas, but they’re held back by your number-driven bullshit. These orgs actively kill innovation by reducing talent to quantifiable rows of data.
A product hobbled together from shit prototype code very obviously stands out. It has various pages that don’t quite look/work the same, Cross-functional things that “work everywhere else” don’t in some parts.
It rewards only the people who make good presentations, or pick the “current hype thing” to work on. Occasionally something good that addresses real problems is at least mentioned but the hype thing will always win (if judged by your SLT)
Shame on you if the slop prototype is handed off to some other team than the hackathon presenters. Presenters take all the promotion points, then implementers have to sort out a bunch of bullshit code, very likely being told to just ship the prototype “it works you idiots, I saw it in the demo, just ship it.” Which is so incredibly short sighted.
I think the depressing truth is your executives know it’s all hobbled together bullshit, but that it will sell anyway, so why invest time making it actually good? They all have their golden parachutes, what do they care about the suckers stuck on-call for the house-of-cards they were forced to build, despite possessing the talent to make it stable? All this stupidity happens over and over again, not because it is wise, or even the best way to do this, the truth is just a flaccid “eh, it’ll work though, fuck it, let’s get paid.”
Maybe, maybe not, it’s hard to tell from articles like this from OSS projects what is generally going on, especially with corporate work. There is no such rhetoric at $job, but also, the massive AI investment seemingly has yet to shift the needle. If it doesn’t they’ll likely fire a bunch of people again and continue.
Not OP, but this is the part that I take issue with. I want to forget what tools are there and have the LLM figure out on its own which tool to use. Having to remember to add special words to encourage it to use specific tools (required a lot of the time, especially with esoteric tools) is annoying. I’m not saying this renders the whole thing “useless” because it’s good to have some idea of what you’re doing to guide the LLM anyway, but I wish it could do better here.
I’ll give it a fair go, but how is it not going to have the same problem of _maybe_ using MCP tools? The same problem of trying to add to your prompt “only answer if you are 100% correct”? A skill just sounds like more markdown that is fed into context, but with a cool name that sounds impressive, and some indexing of the defined skills on start (same as MCP tools?)
> Though you have to watch out for folks that are using the AI to answer your questions.
Heh I do think that happened once (that I was aware of), but it was on a topic I knew a lot about, and it fell apart after layer one. Still, pretty lame, I’d much prefer a “I don’t know,” which I usually say if they start guessing.
That is a really neat interview format, the lightning round of varying themes of common tasks! This right here proves to me you are a good interviewer:
> I'm not worried about whether the string split command takes its parameters in this or that order, I just want to know you know it exists
I’ve run quite a few “can you write code” interviews in the age of practical AI, and I don’t know if I’ve been lucky, am good at breaking through nonsense, or if internet claims are exaggerated, but I can hardly tell the difference between now and the before times. You get someone on a call, you explain a problem, you see how they approach it, you probe along the way. I don’t work for a giant FAANG-like, maybe that’s part of it.
> […] and then having a conversation about what is done, reasoning, etc.
Isn’t this where it would likely unravel?
The interviewer will know what the interesting parts of the exercise are, and ask the deep questions about them. Observe some more: do they know how to use an IDE, run their own program, cut through code to the parts that matter. Basically, can they do the things someone who wrote the code should trivially be able to do?
Since it was mentioned in a sibling comment: Even if the candidate used an LLM to write the code at home, I don’t care, so long as they ace the explanation part of the interview.
> In my opinion the answer is honestly pretty simple: blogs and RSS feeds.
This point is made very often, and I do believe it was true for many people, but I honestly didn’t care about individual blogs at all when I was a young net user.
I didn’t care about the 1,000 words a single person wrote about their trip abroad. There was no way to interact with it? All the action for me was on forums and chat rooms. Like the author mentions, it’s exactly the type of excitement that naturally led to early social media, which I was also a huge fan of for the close friends I already had.
The defeatist in me feels like I will just never have that same feeling again online. In part because I am no longer a child, in part because there are just too many people online now, in part because too many of those people’s brains are twitter-rotted.
It’s fine, I have my close circles to keep my human social spirit alive.
Everything's made up and the points don't matter.
meet.hn/city/39.4064662,-105.4717013/Bailey