michaelbuckbee

FWIW - I did a real world experiment pitting the built in Gemini Nano vs a free equivalent from OpenRouter (server call) and the free+server side was better in literally every performance metric.

That's not to say that the in browser isn't valuable for privacy+offline, just that the standard case currently is pretty rough.

https://sendcheckit.com/blog/ai-powered-subject-line-alterna...

I don't really understand the internal mechanics of of this, but my first thought was why not combine this with a linter/tests. So that it produces all the forks and only keeps the syntactically correct ones.

The underlying intent here (figure out if it's an existing customer of our locally installed apps when they visit our website) doesn't seem bad, but I certainly dislike both the hosts file and localhost detection options.

I'm curious if there's a "good" way to do this.

I build my own products and services and the effective ROI for paying for a more or less unlimited max Claude Code plan is fairly ridiculously positive.

Genuinely asking: has there been a case of the prediction markets being "right" or valuable about something outside of the norm?

Worth reading for the discussion of the LLM watermark technique alone.

Perhaps not so strange in that from a practical standpoint many "AI problems" are actually "Data Problems" when you start peeling things back.

That's a good trick (the "get ready" ping). It reminds me of how early Instagram was considered fast because they did the photo upload in the background while you were typing your caption so that by the time you hit "upload" it was already there and appeared instantly.

It was genuinely quite frustrating that all of the "normal" domain names were taken for this, but not actually used (basically everything "moniker.TLD").

What was surprising was that all the prefix+suffix variants of app/now, etc. were also taken so this was really just me trying to push it hard the other way.

Goofy sideprojects:

https://dnsisbeautiful.com - clean, ad free dns lookup tool.

https://evvl.ai - combination of Github Gists and AI output comparisons (evals)

https://finalfinalreallyfinaluntitleddocumentv3.com/ - free mac app to intelligently rename any kind of file (photos, videos, audio, text) based upon their contents.

Read to the end of the article to learn about NASA's Space Poop challenge and the award winning astronaut diaper the author created.

I'd add AWS + Gravitron to that list as well.

Tahoe introduced some changes to the windowing code that badly disrupted my DisplayPort device that was rock solid on Sequoia. I ended up switching to a new device as a workaround. Window memory use (I have a lot of virtual desktops and extra screens) is much higher and there's a peristent bug where taking a screenshot with CleanShot somehow resets the DisplayPort driver and everything flips out for a minute and it has to rediscover the external monitors. Infuriating.

"signal/noise" possibly?

Well, there are definitely people who care about the vision and style of movies from certain directors. It's not so much "story" like plot, but story in the sense of a "brand story" where there's recognizable elements in all the work, repeated themes, changes and decisions and evolution to how they approach things.

I love using Kindle's as single purpose tablet/interfaces/displays. I'm the weirdo who actually prefers the LCD displays vs eInk and it's incredibly easy to set Kindle Fire's into dev mode which lets you display a webpage, never turn off while connected to power and never show ads.

You can regularly find the Kindle Fire HD10s for ~$40

Addendum to this: my filipina aunt is elderly and I was absolutely shocked at the amount of highly specific AI generated content seemingly targeted directly at her on Facebook.

Except instead of thirst traps it was a weird mix of outrage porn, religious imagery, and kids + pets being cute, singing or rescued from odd situations.

I asked a few questions of her to try and figure out if she like really grasped that it was AI, and she knew the general idea, but there's already so many filters and choppy edits of things it was honestly just too hard for her to make the distinction.

Having picked up a few long neglected projects in the past year, AI has been tremendous in rapidly shipping quality of dev life stuff like much improved test suites, documenting the existing behavior, handling upgrades to newer framework versions, etc.

I've really found it's a flywheel once you get going.

I have a Stream Deck and made a dedicated button for this. So I tap the button speak and then tap it again and it pastes into wherever my cursor was at.

And then I set the button right below that as the enter key so it feels mostly handsoff the keyboard.

I feel the same way, but apparently millions of people are using character.ai?

Kinda sorta the Slate truck - https://www.slate.auto/en

Their pitch is to ship a pretty minimal platform that you can customize up as you want it.

I recently dug into this as I was trying to benchmark the possibility of using Gemini Nano (Chrome's built in AI model) vs a server side solution for a sideproject.

Nano's stored in localstorage with shared access across sites (because Google), so users only need to download it once. Which I don't think is the case with Mistral, etc.

There's some other production stats around adoption, availability and performance that were interesting as well:

https://sendcheckit.com/blog/ai-powered-subject-line-alterna...

Been making a few different sideprojects:

1. https://app.evvl.ai - a privacy focused Eval tool that you can download and compare outputs from different AI models (bring your own key)

2. https://dnsisbeautiful.com - an extremely clean and fast DNS lookup tool to track propagation across the different networks.

3. https://finalfinalreallyfinaluntitleddocumentv3.com/ - this is a Mac only file renaming tool (you drag and drop "screenshot1.png" onto it and it renames it to "dns-results.png" for you via local AI models. Though in true side project fashion the site looks kind of borked at the moment.

I'm now letting Claude Code write commits + PRs (for my solo dev stuff), but the benefits have been pretty immense as it's basically Claude keeping a history of it's work that can then be referenced at any time that's also outside the code context window.

FWIW - it works a lot better to have it interact via the CLI than the MCP.

I think this is probably less effective than if there was some sort of "credit" or reputational score for reporting that seems like something GitHub would have the information to implement.

I have two kids (sophmore in HS and a middle schooler) and in both their individual studies and when I'm helping them with homework we use AI pretty extensively now.

The one off stuff is mostly taking a picture of a math problem and asking it to walk step by step through the process. In particular this has been helpful to me as the processes and techniques have changed.

It's been useful in foreign languages as well to rapidly check work, and make corrections.

On the generative side it's fantastic for things like: give me 3 more math problems similar to this one or for generating worksheets and study guides.

As far as technological adoption goes, it's 100% that every kid knows what ChatGPT is (even maybe more than just "AI" in general). There's some very mixed feelings from the kids with it: my middle schooler was pretty creeped out by the ChatGPT voice interface for example.

I still find it to be the case that most 100% cotton shirts shrink over time (even pre-shrunk) and have switched to blends just to get some more longevity out of them.

Unless you hit a performance wall with Postgres or absolutely need Batch capability you've probably got a very large runway with SolidQueue.

I wrote this article about migrating from Redis to SQLite for a particular scenario and the tradeoffs involved.

To be clear, I think the most important thing is understanding the performance characteristics of each technology enough that you can make good choices for your particular scenario.

https://wafris.org/blog/rearchitecting-for-sqlite

FWIW, Sidekiq docs strongly suggest only passing around primary keys or identifiers for jobs.