Відео

Thanks for making this video. I've just tried OLLAMA. It gave wrong answers to 3 of my 4 first (and only) prompts. Uninstalled it.

I have installed wasmtime, wasmedge runtime in k8s node. how can i access files to wasm module with k8s? for the normal hello-world application, I created OCI image as per below file. FROM scratch ADD hello-wasm.wasm / CMD ["/hello-wasm.wasm"] since wasmtime, wasmedge runtime is running on my node so this is working. but how can i access file system? do you have any reference tutorial or github link for this type of implementation?

Is there any possiblities to fine-tune, quantise or distill the model into a smaller model to be run locally? I urgently need this for a project I'm doing.

igrore the dude below, awesome content , keep going .

It's so fucked up that the American pronunciation of quay is "kway", lmao.

we’d love the longer version

Great video! I'm the PM for TypeSpec at MS, from the comments it looks like there is interest in back-end code generation from TypeSpec, which we're currently working on.

Maybe you need new glasses. Your eyes keep getting stuck together, hella annoying.

3:00 - Show how to do it on a local machine please

For example, RPN for stack is like the following: Like doing math vertical when in school 100 (first number entered... Bottom of Stack) 10 (Second Number entered (Higher up in stack or on top in this case)) - ___________ ( Do the mathematical operation: Subtract ) Hope the format remains.

Thanks for the video series. I think of WAT2WASM more as a "web assembler" translating "WAT" (WebAssembly textual source) to (WebAssembly Virtual Machine Language [ VML ] akin to P-Code) and high-level languages like Javascript (web "BASIC" or web "Pascal"). This is because WAT2WASM is more like Assembly to Machine Language translation versus compilers which are higher up in the totem from Low, to Intermediate, to High-level programming languages. Assembly language for various native CPUs were considered intermediate-level languages. A step closer to the "bare metal" of machine language than C or Pascal or BASIC were at source. BASIC, albeit were commonly interpreted, did have compilers that existed. On 8-bit computers like C64, we had many. Several that compiled to P-Code which was a VM bytecode thing much like this is. Not quite ML speed but considerably faster than interpreters and gave performance boosts. It is unclear how the speed is compared to JIT compiled JS once it had been ran through JIT compiling but this is a binary that is platform independent than native binary of x86 or ARM or whatever.

I felt cursor was better than continue but yes continue is open source, cursor is not.

I have a weird issue when i open a folder continue crashes. I have been using deepseek 6.7b for fill in the middle. It works well.

you need more views, great content

Thank you! This is gold

the blue blocking glasses with the high intensity blue lights in the background makes this video

I think it doesn't provide some important features like inline chat

Amazing one! Thanks very much for this. Instead of Ollama, do you think is it doable through Llama.cpp ? Not everyone org allows to install Ollama 😢

first like :)

Outstanding! Did you try this approach with Llama3, Llama Instruct, Code Llama, StarCode or Deep seek? Thanks, you have the best tutorial in this topic but the result is no good enough yet ;)

Thanks😇

5:30 I have a problem. I'm running Surreal version 1.5.2 for Windows on x86_64. Whenever I start a server with "surreal start", regardless of any optional commands, the command line interface stops accepting input. It seems to start a server, but it leaves me with a blinking underscore, and I can't type anything more after that. Am I missing something here?

This was so helpful, thank you!

That your normal lighting setup?

How is this possible, looks too good to be true

Too much effort. Just use containers. How would you package a tool that has tonne of dependencies using web assembly?

Chris, I really enjoyed the video thanks. My world is C/C++ and Java and recently Python. I managed for follow the video well but was lucky I had the node stuff installed on my laptop. What would be interesting to me is how to now implement a simple back-end service in Python (or Java) that exposes the APIs you have defined and something simple (out of my knowledge base) on the front-end to interact with the API. My suggestion would be something like a very simple chat-response from a local LLM.

Excellent explanation. Thanks for making these examples.

Interesting, thanks.

Bravo! 🎯

The most useful and informative video I saw on this topic. Appreciate❤

Hello Chris! Do you have an idea what i need to do in VSC to see the real time Swagger on the side interface? Saw this a lot of times but don't know how to do it

Did you mean "function calling" at 10:10 ?

Whitespace significance does my head in, otherwise Nim is indeed interesting.

Having written spec manually in order to use with Swagger, I can see that it is a pain to do, but I did so in JSON to avoid YAML & I think that is a lot easier (why do we need YAML, when TOML exists.. but anyway) If this could then create the routes for me, it would be a significant benefit, but otherwise I am learning all those decorators, various imports, namespaces and using statements, having to compile for each change etc, when I could have just written myself manually, right.. Good video, even if I'm not convinced on it ☺️

Awesome video. The question is now how to use the typespec to implement typesafe apis in typescript without having to rewrite the spec in typescript

Not Alan Turing invented the stack, but F.L. Bauer and Samelson invented it independently in the 1920s. -> en.m.wikipedia.org/wiki/Stack_(abstract_data_type)

Damn, thank you UA-cam for recommending this channel. @chrishayuk is a gun. Thanks Chris

I am only new to rust, so this is quiet helpful. Do you have any examples where you send custom headers? As you would do when you need to authenticate. My attempts have been halted by: Trait `IntoClientRequest` is not implemented for `Result<Request<()>>

Hey man! Super helpful. Do you have any thoughts on how this can be useful for agents that need to create code and ensure it works?

clean, and crisp! brilliant video chris :) PS: Can you please update the tokenizer part of the code, it's a bit confusing, thanks!

With rust no one gets jobs

Register count is mentioned as 0 in the generated bytecode. Instruction Ldar a1, loads whatever content is available in register a1. Then register count should be 1. Can you please explain?

Hello, I was looking at your video channel. We may be helping a company that uses secure images to increase supply chain security and help cloud native development. Would you be willing to help try their software, make a video, and help show devs how to use their tools? This is not an offer, but just to start a conversation about your willingness to take on sponsorship. Please provide me with your email if you are interested. You'd have a chance to look at their technology and decide if it's the type of software that you'd be interested in covering in your channel.

Very good, very clear 👍

how to use big.js with assembly script

and it's faster than crystal, so i slightly more like it. except the syntax, i more like ruby-like syntax rather than nim (pascal-like).

The music is loud and the volume is chaotic in the intro

So this is what prompt engineering really is

You really need to explain why models ability has nothing to do with its model size. It has to do with the human’s ability to guide those models. GPT went through that already. Now you do it with a new model and boom