senko

37 ms·

721.
by senko 2y ago
Go build one? (with or without YC support) I do like your comment as the critique of contemporary tech entrepreneurship culture that is very hype oriented, likely at the expense of comparatively few blue ocean ideas that get crowded out. Ho
722.
by senko 2y ago
Yes. 250ish per cohort meant the partners were spread very thin, and founders could realistically connect with just a subset of batchmates. I'm glad they're scaling back the batch size.
723.
by senko 2y ago
Others have given good answers, but what I meant here was you got to have a certain appetite for risk, (naive?) confidence in your abilities, few family/monetary obligations, willingness to go to the extremes, and at least a bit of dis
724.
by senko 2y ago
My feeling is that if you think quitting the day job is risky, you probably won't fit the profile YC and VCs look for. YC is an enormous brand in the VC and startup industry, and as a YC startup doors open to you that would be harder t
725.
by senko 2y ago
One of the most impressive companies in my batch (W24) had nothing to do with AI, or indeed software: https://www.astromecha.co/ So no, not a requirement. OTOH if you do come with an AI idea, they won't reject you just
726.
by senko 2y ago
The email can have a video link. Unless there's a need to interact (and there rarely is at an all-hands), it would save everyone time to just prerecord it. Any followup questions can be sent via email with reply-to-all (or an internal
727.
by senko 2y ago
> Why would anyone ever publish stuff on the web for free unless it was just a hobby? So that ChatGPT mentiones you , not your competitor, in the answer to the user. I have seen multiple SEO agencies already advertise that.
728.
by senko 2y ago
This nails it: > Who Gets To Decide What Is True? I would not agree to this however: > It was then up to the reader to decide which of those sites they wanted to visit, With current search engines, Google decides for you ("helped
729.
by senko 2y ago
There are so many solid business ideas that take VC money, turn out not to be unicorn potential, and crash and burn, where a slower, sustainable growth might work just fine. Gumroad is a famous example.
730.
by senko 2y ago
Apparently that's Codium, who have recently renamed themselves to Qodo: https://www.qodo.ai/blog/introducing-qodo-a-new-name-the-sam... (TIL)
731.
by senko 2y ago
VSCode uses ordinary ssh tho. I multiplex my ssh connections so the workflow is just ssh sgain, then reload VSCode window. If Mosh could multiplex these (and paper over the connection problems) that'd be great but after a cursory look,
732.
by senko 2y ago
There's no input delay in VSCode (editor, ui) because the UI is local. Delay in saving/reading/sesrching in files is not noticable for me. (edit to explain: VSCode is still running locally, but it also installs the server-sid
733.
by senko 2y ago
I use VSCode + SSH remote for this and works great. The only nitpick I have is needing to manually reconnect when I suspend my laptop and ssh connection breaks. It's a separate session though, which doesn't matter to me but may be
734.
by senko 2y ago
Kagi user here, I almost never need to !g. Used DDG a few years back and stopped after I had to fallback to Google for most qs.
735.
by senko 2y ago
Subscribing to Kagi is definitely easier (and less costly!) than handling the pandemic, ensuring world peace and utopian government. Edit to add: and since we're talking about indie web: https://search.marginalia.nu/
736.
by senko 2y ago
Curious, I would expect a programmer of your age to remember Knuth's "beware of the bugs in above code, I have only proven it's correct but haven't actually run it". I'm happy you know math, but my point before
737.
by senko 2y ago
I didn't say it's hard, but it's most definitely leetcode, as in "pointless algorithmic exercise that will only show you if the candidate recently worked on a similar question". If that doesn't satisfy, here&#x
738.
by senko 2y ago
We collectively deride leetcoding interviews yet ask AI to flawlessly solve leetcode questions. I bet I'd make more errors on my first try at it.
739.
by senko 2y ago
By the same arguments: People who live can also be divided into groups who would have and who would not have paid for a game. There is damage done, despite all the people who only live ethically. Even the people who do something else (than
740.
by senko 2y ago
Even if that were true (and I'd challenge that assumption[0]), there's no dichotomy here. Software quality, for the most part, is a cost center, and as such will always be minimal bearable. As the civil engineering saying goes, an
741.
by senko 2y ago
> Companies' need for people grows quadratically in proportion to the amount they are trying to do, not linearly. That's fine. The question the poster raises is how come they have 500 over the number of people they need, if the
742.
by senko 2y ago
Other Fedora 41 spins are released too, here are the notes for the GNOME ("default") version: https://fedoramagazine.org/whats-new-fedora-workstation-41/
743.
by senko 2y ago
The part you cut from the quote actually gives it the opposite meaning: emulation was demonized so only the people who were okay with piracy considered it. Going from that to "emulation caused irreparable harm to developers" takes
744.
by senko 2y ago
Tell me you're an IP lawyer without telling me you're an IP lawyer: > Here Cifaldi clearly acknowledges that emulation is a powerful means of piracy that has done irreparable damage to the intellectual property rights of game d
745.
by senko 2y ago
A different AI toolbox, for people interested in results, not hot stuff. My pet theory is that the popular libraries and frameworks we have today (LangChain, LlamaIndex) are first generation products, where just getting the damn thing to wo
746.
by senko 2y ago
My rule of thumb for “reduced X” or “no Y” labels is to assume the (food) product’s got loads of everything else, and that heuristic rarely lets me down.
747.
by senko 2y ago
Error parsing your reply: TypeError: this is a haiku but a limerick was requested. Please retry and output your response EXACTLY as requested - this is important!
748.
by senko 2y ago
You jest, but I believe @tptacek is using an LLM (ChatGPT?) to understand the details of various Linux kernel subsystem and has said it works quite well for the task. It's not a great jump from that to "port Linux device driver fo
749.
by senko 2y ago
All languages have their warts. In JavaScript, the warts have their language.
750.
by senko 2y ago
The problem with using the current crop of LLMs for coding, if you're not a developer, is that they're leaky abstractions. If something goes wrong (as it usually will in software developent), you'll need to understand the und

More ›