tymscar

So basically the best way to use MCP is not to use it at all and just call the APIs directly or through a CLI. If those dont exist then wrapping the MCP into a CLI is the second best thing.

Makes you wonder whats the point of MCP

Love the idea, but please add some demo screenshots on GitHub. All UI tools should

https://blog.tymscar.com is my blog. I write mostly about technology and programming

Are they multipliers when you do less of it and offload more of it to the same tool everyone else uses?

I agree, however there you don’t compare technical knowledge alone, you also compare managerial skills.

With LLMs its admittedly a bit closer to doing it yourself because the feedback loop is much tighter

I do think your datasettes work is fantastic and I genuinely hope you take my previous message the right way. I’m not saying you do something bad, quite the opposite, I feel like we need more of you and I’m afraid because of LLMs we get less of you.

I think you are right in saying that there is some deep intuition that takes months, if not years, to hone about current models, however, the intuition some who did nothing but talk and use LLMs nonstop two years ago would be just as good today as someone who started from scratch, if not worse because of antipatterns that don’t apply anymore, such as always starting a new chat and never using a CLI because of context drift.

Also, Simon, with all due respect, and I mean it, I genuinely look in awe at the amount of posts you have on your blog and your dedication, but it’s clear to anyone that the projects you created and launched before 2022 far exceed anything you’ve done since. And I will be the first to say that I don’t think that’s because of LLMs not being able to help you. But I do think it’s because what makes you really, really good at engineering you kept replacing slowly but surely with LLMs more and more by the month.

If I look at Django, I can clearly see your intelligence, passion, and expertise there. Do you feel that any of the projects you’ve written since LLMs are the main thing you focus on are similar?

Think about it this way: 100% of you wins against 100% of me any day. 100% of Claude running on your computer is the same as 100% of Claude running on mine. 95% of Claude and 5% of you, while still better than me (and your average Joe), is nowhere near the same jump from 95% Claude and 5% me.

I do worry when I see great programmers like you diluting their work.

Its one of the main reasons why I buy on steam. Makes games much more engaging, especially for me because I prefer hard action games with little to no story.

Honestly it’s very much worth it! Look into this https://hexdocs.pm/hexdocs_offline/index.html

Yeah, so DDR5 is 8GT and GDDR7 is 32GT. Bus width is 64 vs 384. That already makes the VRAM 4*6 (24) times faster.

You can add more channels, sure, but each channel makes it less and less likely for you to boot. Look at modern AM5 struggling to boot at over 6000 with more than two sticks.

So you’d have to get an insane six channels to match the bus width, at which point your only choice to be stable would be to lower the speed so much that you’re back to the same orders of magnitude difference, really.

Now we could instead solder that RAM, move it closer to the GPU and cross-link channels to reduce noise. We could also increase the speed and oh, we just invented soldered-on GDDR…

DDR5 is a couple of orders of magnitude slower than really good vram. That’s one big reason.

Thank you! I have not played around with Elixir just yet, sadly, so I can't help there, I'm afraid.

Thank you! TIL

You're right, but loads of times I just left that there because I probably did something more involved in the map that I ended up deleting later without realising.

You most likely asked an AI for this. They always think there is an `if` keyword in case statements in Gleam. There isn't one, sadly.

EDIT: I am wrong. Apparently there are, but it's a bit of a strange thing where they can only be used as clauses in `if` statements, and without doing any calculations.

I was also pretty surprised about the performance. It's not C, but it's so much faster than I'd guessed.

I do agree the language server is great. And it works in basically any IDE, which is another huge bonus.

With regards to having to type `list.map`, you actually don't need to! You can do this:

  import gleam/list.{range, map}
  import gleam/int

  pub fn main() {
    range(0,10) |> map(int.to_string) |> echo
  }

Some libraries just aren't there, and I do wonder how hard it would be to port C libraries over. Something I want to play with!

I actually agree. I don’t even use them in my private code. I will look into removing them from my blog!

Hey. I’m glad you got it working. I didn’t intend it to be a step-by-step tutorial, more like a “hey, this is how I solved it”, with some details like the configs in case you want to follow along. And yes, the website itself still does not work, no matter what URLs you try to fix; it’s just the images, which, for me, are good enough.

I guess at the end of the day it was a tradeoff we had to take for the lack of a better alternative back then

Oh, I guarantee you that this has not been touched by any AI. I used to use emdashes all the time, then people thought those were AI telltale signs, so I stopped. I loved making lists. Same thing.

Now I’m not allowed to say “key detail”??

That’s what I get if I go to their page, but when curling the image directly it works fine

Do you people think nobody calls your bluff?

Here’s the launch card of the sonnet 3.5 from a year and a month ago. Guess the number. Ok, Ill tell you: 49.0%. So yeah, the comment you replied to was not really off.

https://www.anthropic.com/news/3-5-models-and-computer-use

It can never be perfect because I can’t even read my handwriting. Good thing the need for it dropped to basically 0 as soon as I finished school.

Here’s one where I did a Gaussian Splat. But I talk about my LiDAR scanned head there too

https://blog.tymscar.com/posts/gaussiansplat/

As it turns out, I do play with LiDAR all the time. I think I have a blogpost about scanning my environment too

The maximum routing speed Unifi Dream Router 7 can do with IDS on is 2.3Gbps according to their spec sheet

Oh, good point. That was actually the first thing I missed, but when I created a new Wi-Fi 7-only SSID, Unifi wouldn’t let me pick anything lower than WPA3 if I only used 6 GHz. So that sort of fixed itself.

But also faaaaaaar slower

Running ethernet to every room is always going to be a good idea

All the replies you get here are totally valid, but Ill throw in another one.

Why not? Life’s too short anyways, and playing around with tech is one of those things that bring me joy.