gptcli VS nushell

You must have an account with Open AI. If you don't have one, go here and create one.

For a long time, I was looking for something exciting to work on and this was it. The fact that ChatCraft already supported Speech to Text transcription using Whisper, which is another one of OpenAI's models with unique capabilities, integrating Text-to-Speech would essentially turn our application into something like an Amazon Alexa but with a brain powered the same LLM that ChatGPT uses.

With your API key, you’ll be able to make authenticated HTTP requests to OpenAI. So it’s a good idea to get familiar with the API itself. I’d encourage you to take a brief look through the OpenAI Documentation and become familiar with some concepts. The models are particularly good to understand because they have varying capabilities.

Next, we’ll also need an OpenAI account to have access to their API.

Capable of reasoning:The application can be connected to a large language model like OpenAI, Cohere, or Hugging Face. This enables the application to answer based on the context it is given.

Beyond that, "AI tools" is a pretty broad category. Odds are you're asking about LLMs (perhaps even specifically GPT), in which case a good place to start is with OpenAI and their API. Alternatively, Hugging Face is a good place to find models (not just LLMs, either) and they have tooling to let you interact with them in a consistent way. In these cases, you're not actually developing an AI, per se, but just utilizing it. But when I think of "AI Tools" at this point in time, I'm thinking of tools built on top of LLMs.

Like most things in IT and programming, this imo is also best learned by doing rather than reading a bunch of technical papers. I recommend starting with ChatGPT and testing out varieties of prompts asking it to fill in your knowledge. There are great documentation by both OpenAI and Microsoft Azure on using these models with examples. For open-source models (and transformers in general) HuggingFace is a great resource and you can get started on some smaller models by downloading it and trying right away. Karpathy's nanoGPT video on Youtube is also very useful to get started.

It's just a gui to (most of) what you get through the API. Read the API docs for details of each option: https://platform.openai.com/docs/introduction

For the full list of options, check out the OpenAI API documentation.

If you are interested in an API for ChatGPT in general, you can go look at the documentation or this starter guide for gpt-4 and gpt-3.5-turbo on the openai website. This should then be adapted to the language you are using.

The first method is through downloading the pre-built binaries. With this method, you don't need to install anything other than Nushell's dependencies. Once you've downloaded the binaries, add them to your system's environment path to run it directly in your terminal.

I rather nushell for this purpose, it's more fun to write and easier to read.

https://www.nushell.sh/

These are just three small examples of what this shell written in Rust allows. The features are many and many more, but I'll leave it up to you to discover and enjoy them; I'm currently playing around with it and it's giving me a lot of satisfaction and immediacy, now it has a fixed place among the tools I use when working! The project is Open Source, so if you want to contribute, I invite you, as always, to do so, I leave you the link to the repo here!

Any thoughts on fish as compared to nushell [0]? It's similar to PowerShell in its philosophy and is also written in Rust.

[0] https://github.com/nushell/nushell

> In PowerShell, structured output is the default and it seems to work very well.

PowerShell goes a step beyond JSON, by supporting actual mutable objects. So instead of just passing through structured data, you effectively pass around opaque objects that allow you to go back to earlier pipeline stages, and invoke methods, if I understand correctly: https://learn.microsoft.com/en-us/powershell/module/microsof....

I'm rather fond of wrappers like jc and libxo, and experimental shells like https://www.nushell.sh/. These still focus on passing data, not objects with executable methods. On some level, I find this comfortable: Structured data still feels pretty Unix-like, if that makes sense? If I want actual objects, then it's probably time to fire up Python or Ruby.

Knowing when to switch from a shell script to a full-fledged programming language is important, even if your shell is basically awesome and has good programming features.

Maybe if the "popular" shells, but http://www.nushell.sh/ is looking better and better

Yeah agreed, especially now that PowerShell is available cross-platform.

Nushell[1] also seems like a promising alternative, but I haven’t had a chance to play with it yet.

[1]: https://www.nushell.sh/

I also discovered an existing discussion[1] related to this topic which includes a link[2] to a "helper to call nushell nuon/json/yaml commands from bash/fish/zsh" and a comment[3] that the current nushell dev focus is "on getting the experience inside nushell right and [we] probably won't be able to dedicate design time to get the interface of native Nu commands with an outside POSIX shell right and stable.".

[0] https://gitlab.com/RancidBacon/notes_public/-/blob/main/note...

[1] "Expose some commands to external world #6554": https://github.com/nushell/nushell/issues/6554

[2] https://github.com/cruel-intentions/devshell-files/blob/mast...

[3] https://github.com/nushell/nushell/issues/6554#issuecomment-...