HN user

collinvandyck76

456 karma

https://5xx.engineer

[ my public key: https://keybase.io/keys_so_fresh; my proof: https://keybase.io/keys_so_fresh/sigs/pAawFAOZimt61ZRuBydZW39Vmap-qR_siBNC06L8PBw ]

meet.hn/city/us-Durham

Posts1
Comments196
View on HN

Most are, implicitly at least. But at the time, there was definitely a period of transition with many shops continuing to use subversion for a while.

I think about this a lot, and do everything I can to avoid having Claude write production code while keeping the expected tempo up. To date, this has mostly ended up having me use it to write project plans, generate walkthroughs, and write unit and integration tests. The terrifying scenario for me is getting paged and then not being able to actually reason about what is happening.

This is an odd comment on "Hacker" news. There's a joy to be had in making things like this, even though there are better alternatives and it's often a weird thing to do in the first place. I often do not post the things I create here and elsewhere because I dread this kind of feedback. What do you hope OP takes away from your post?

There were a number of Motorola-inspired ideas that made their way to Dilbert while I was working there in the late 90s.

This is true, also for people immersed in this world. Sometimes the dark mode of a site is ass, and it's better to set a preference for that site to use light mode to make it more usable.

Always good to have a weed puller in your toolshed. A stand-up puller, specifically, that operates as a lever, allowing it to first grab deeply and then through a rotation of the handle it pulls out quite a bit of the root system. A lifesaver if you have a rain garden which is really just a synonym for weed garden.

Yeah, this. Each project I work on has it's own markdown file named for the ticket or the project. Committed on the branch, and I have claude rewrite it with the "current understanding" periodically. After compacting, I have it re-read the MD file and we get started again. Quite nice.

Claude Opus 4.1 12 months ago

Claude Code is the superior interface in my opinion. Definitely start there.

My story is similar to yours, and I feel what you're saying. I would just be a different person today if I had avoided it altogether. Internally I have resolved to just keep looking forward, and only very occasionally, backwards.

It sounds kind of anachronistic, although I think there might be some markets for that outside of this bubble. If you're looking for critique, I think your landing page is extremely noisy for such a simple idea. You'd probably get better traction with a more focused design.

I finally got access to it -- much better experience than I had expected. I really like being able to choose the voice/cadence. I wasn't able to get it to auto-send after a few seconds, but that might have been due to being in a noisy environment.

I honestly used to be a skeptic on AI, at least for how it was being marketed, until agentic workflows became available. I've been in this biz for about 25 years now and have seen a lot of fads come and go, but framed correctly, this one has legs. If you're using it as a foundational part of your development budget I have bad news for you. But if you're using it as a handy assistant that you can guide to do things in minutes that would take you hours it's a force multiplier.

One interesting thought I had about it after I used Claude Code to generate a lot of documentation for a new codebase I was being bootstrapped into was that it's kind of a hedge against the loss of mental acuity that is natural with aging. I, and friends of mine, have admitted that you kind of get less quick/sharp with age. The tradeoff here is wisdom. But with a tool like this I'm able to move much more quickly than before.

Thanks a lot. This sounds great for my envisioned use case which is having conversations while on a road trip to research ideas that pop into my head. Not having to tap the screen makes it much safer.

Thank you for sharing. I'm looking forward to getting voice mode, but after doing some searching I saw that it still needed the user to hit the "send" button after they are done talking. Did I misread, or was your experience that it would reply to you after you finished talking without the need to press anything?

Stepping Back 1 year ago

I materialize my thoughts as I obsess over the problem in my notes. I find that doing this often helps me reframe the problem and the facts that I currently know, which helps me discard branches that will be fruitless.

I'm a Collin with two l's. Not as uncommon as Gregg but it's never been a burden. I also don't bother correcting people unless it's happening a lot with the same person and someone I expect to interact with a lot.

I gave it a shot a couple years ago -- pretty slick. But the latency was noticeable compared to tmux, which I remain using today. I admit that I'm kind of sensitive to it, as at the time I was already on a latent connection.