It can't function without advertising, money, or oxygen, if we're just adding random things to obscure our complete lack of an argument for advertising. We can't go back to an anaerobic economy, silly wabbit.
- Made a RAG in ~50 lines of ruby (practical and efficient)
- Perform authorization on chunks in 2 lines of code (!!)
- Offload retrieval to Algolia. Since a RAG is essentially LLM + retriever, the retriever typically ends up being most of the work. So using an existing search tool (rather than setting up a dedicated vector db) could save a lot of time/hassle when building a RAG.
I built a similar system for php and I can tell you what is the smart thing here: accessing data using tools.
Of course tool calling and MCP are not new. But the smart thing is that by defining the tools in the context of an authenticated request, one can easily enforce the security policy of the monolith.
In my case (we will maybe write a blog post one day), it's even neater as the agent is coded in Python so the php app talks with Python through local HTTP (we are thinking about building a central micro service) and the tool calls are encoded as JSON RPC, and yet it works.
I had to do something similar. Ruby is awful and very immature compared to python, so I "outsourced" the machine learning / LLM interaction to python. The rails service talks to it through grpc / protobuf and it works wonderfully.
While I agree that the training/learning ecosystem is pretty heavily centered in Python, going from that to "Ruby is awful" seems like a very drastic jump, especially if we are talking about the LLM interaction only.
I probably wouldn't write a training system in Ruby (not because it's not doable, just because it's not a good use of time to rewrite stuff that is already available in python ecosystem)... but hooking up a Ruby system up to LLM's for interaction is eminently doable with very little effort.
I am assuming your situation had some specific constraints that made it harder, but it would be nice to understand what they were - right now your comment describes a more complicated solution and I am curious why you needed it.
While I agree that Python is where most of the implementation action is, one of the great things about building applications with LLMs is that almost all API providers offer a rich REST interface, and I have found it simple to use LLM services in Haskel, various Lisp languages, etc. It is nice having very old code in various languages and be able to add new functionality with LLMs.
By the guidelines as written, it isn't. By the guidelines as mysteriously and generously interpreted by the hall monitors, it's the most beautiful thing to ever exist on God's green earth. Nothing says "hacker spirit" like waffle-stomping AI into software that was already working just fine.
Ok, if you have such insight into development, why not leverage agents to type for you? What sort of problems have you faced that you are able to code against faster than you can articulate to an agent?
I have of course found some problems like this myself. But it's such a tiny portion of coding I really question why you can't leverage LLMs to make yourself more productive
Most coding is better done with agents than with your hands. Coding is the main financial impediment to development. Yes, actually articulating what you want is the hard problem. Yes, there are technical problems that demand real analytical insight and real motivation. But refusing to use agents because you think you can type faster is mistaking typing for your actual skill: reasoning and interpretation.
Windows is reasonably OK, but MacOS' window management has always been really terrible.
Just think through the many different iterations over the years of what the green button on the deco does, which still isn't working consistently, same as double-clicking the title bar. Not to mention that whatever the Maximize-alike is that you can set title bar double click too (the options being Zoom and Fill, buried in settings somewhere) is different from dragging the title bar against the top of the screen and chosing single tile. Which is different from Control-Clicking the green button. Maybe. It depends on the app.
What a mess.
Both of them miss (without add-ons) convenience niche features I cherish, such as the ability to pin arbitrary windows on-top, but at least the basics in Windows work alright and moreover predictably and reliabily. Window management in MacOS just feels neglected and broken.
There may be many other ways in which MacOS shines as a desktop OS, and certainly in terms of display server tech it has innovated by going compositing first, but the window manager is bizarrely bad.
There is at least one area where both macos and windows suck - handling window focus. MacOS is regularly having trouble with tracking focus across multiple monitors and multi-window apps, making it unusable with keyboard only. And Windows just loves to steal focus in the most inappropriate moments.
>> windows and macos cant even do proper window managing for a start
> Well they certainly manage them better than x11 and wayland.
X11 doesn't manage Windows. You'd know this if you used it, and if you've used it, you'd know why some consider the window management on Windows and MacOS very primitive.
> X11 doesn't manage Windows. You'd know this if you used it, and if you've used it, you'd know why some consider the window management on Windows and MacOS very primitive.
Sure. Windows and macos are also fallible. But there has never been a project that competes with these two brands that can boast a similar commitment to stability and usability.
I don't use a Mac, but have you ever used Windows?
I mean, maybe you have, but if you are not fussy then at worst MacOS is quirky and Windows and Linux are identical and merely have different icons.
If you pay a little bit of attention you will notice that on linux things seem more flexible and intuitive.
If you are very finnicky, there is nothing that comes close to X11 window managers when it comes to window management flexibility, innovation and power.
Windows allows you to launch applications from a menu or via search. You can switch between windows with a mouse or keyboard shortcuts. Windows can either be floating, arranged in pseudo-tiled layers, or full screen. KDE can pretty much do the same under Wayland. Ditto for Gnome under Wayland, albeit to a lesser degree. That covers the bases for most people.
X11 window managers were a mixed bag. While there were a few standouts, most of the variation was in the degree to which they could be configured and how they were configured. There may be fewer compositors for Wayland because of the difficulty in developing them, but the ones that do exist do standout.
> I don't use a Mac, but have you ever used Windows?
I have
> I mean, maybe you have, but if you are not fussy then at worst MacOS is quirky and Windows and Linux are identical and merely have different icons.
Neither have keybindings that make any sense. The other failures are secondary
> If you pay a little bit of attention you will notice that on linux things seem more flexible and intuitive.
Only for windows refugees that have never used Mac OSX
> If you are very finnicky, there is nothing that comes close to X11 window managers when it comes to window management flexibility, innovation and power.
Unless you want to copy and paste, or have consistent key bindings cross applications, or take screenshots. Sure
> European center and left parties could suck all the oxygen out of the room and starve the far-right overnight if they simply introduced and enforced major immigration restrictions
Economic suicide. Why would anyone argue for this? Europe might as well just nuke itself.
Granted, if I were a conservative European, I would also be pro nuking myself.
reply