mui-toolpad VS hamilton

This seems to mainly be useful for spinning up quick and dirty internal tools.

But for that use-case, isn't it easier to use something visual and established like Retool (https://retool.com/) or that generates nice react code, like MUI Toolpad (https://mui.com/toolpad/)?

How does it stack up against MUI's Toolpad? (https://mui.com/toolpad/)

All things considered, they seem pretty similar - visual UI to generate React code that works alongside existing codebase, open-source & self-hostable, etc.

- All configuration is stored in local files which you can version-control, edit, and deploy in any way you want.

You can check out our live demo [1]. If you find it useful, you can support us by giving a star on GitHub [2]. We released our public beta [3] this week. We are happy to answer any questions/feedback in the comments.

[1]: https://stackblitz.com/fork/github/mui/mui-toolpad/tree/mast...

[2]: https://github.com/mui/mui-toolpad

[3]: https://mui.com/blog/2023-toolpad-beta-announcement/

- Vercel for hosting, because they take a Git repo and host it for you in a couple clicks and manage everything. Free or cheap ($20/mo) at MVP stage.

- Next.js (Vercel's open-source React framework) will handle frontend tooling, routing, type checking, and linting for you with a single command (`npx create-next-app`). Starting the server is one more command (`next dev`) and your page is up and running.

- For the UI layer, I'd recommend either starting with one of their prebuilt templates (https://vercel.com/templates/next.js) and modifying it as needed

OR using a modern component system like https://mui.com/ or https://ant.design/ or https://chakra-ui.com/ instead of trying to learn and write your own component and JS+CSS code. Using one of these systems will allow you to compose complex apps out of well-made, well-documented, easy-to-use primitives, making it much easier to focus on business needs rather than basic frontend components and infra.

The basic MUI system, for example, is totally free. You can find third-party apps built on top of it (https://mui.com/store/#populars) and pay a one-time license fee to essentially "fork" them, getting a prebuilt working app that you just attach your backend API calls to.

There are also low-code extensions of these frameworks (meaning you start with a GUI, plan out your app that way, but still have access to the source for future advanced changes). Examples are https://mui.com/toolpad/ and https://retool.com/use-case/dashboards-and-reporting

----------------------

Is this a lot? Yes and no. React has a learning curve of its own, but it can take the place of having to learn raw HTML and CSS. (Yes, you eventually should know those things for debugging and polishing, but they are largely a level of abstraction below what you really need for a basic MVP).

Once you learn React, its primary value isn't that it's a great language (opinions differ) but that it has a humongous ecosystem of third-party vendors, free open-source libraries (basically any component you might think to build is probably already available on npm), and a wide availability of devs from hobbyists to full-timers.

Others in this topic will suggest going away from Javascript as much as possible (and using things like HTMX or backend-to-HTML solutions like the old days). That's fine, but you lose out on the rich ecosystem of React and Javascript, so you end up having to build more yourself -- which is what you're trying to avoid in your case.

My own 2¢: As someone who grew up with HTML and made websites since the birth of Javascript and CSS, the web has always been messy. It's always been a semi-open ecosystem controlled by a few major companies (whether that's Netscape or Microsoft or Sun or Adobe, or these days Google and Apple), so it very much suffers from design-by-bullying. Whoever is the power player of the decade gets to add their favorite technologies that everyone else is forced to adopt. Thus the web became a hodgepodge of document markup systems poorly fitted for modern apps, with various hacks on top of hacks built to satisfy some big company or another's in-house needs. Sadly, that means going "vanilla HTML+JS" doesn't leave you with much, just the shattered legacy of poor historical decisions.

React at least helps by encouraging componentization and abstraction of UI elements to functions, using cleaner data models (actual variables and objects) vs direct DOM manipulation (storing page content as state).

We've gone through many generational shifts in approach, from the raw HTML days of Geocities to the you-build-it, we-host-it approach of Godaddy and its ilk, to the "all in one" CMSes like Wordpress or Drupal. These days, (if you want there to be), there can be a pretty clear separation between backend and frontend systems, and with that specialization came a bunch of startups (mentioned above) whose approach is "let us help you build it as best as we can, so you can focus on business logic instead of basic UI and infra". After 20 years of doing this, the current state of the web developer experience is actually my favorite so far. HTML and CSS suck for building apps (as opposed to documents), and although Javascript is a lot better since ECMAscript v6 (ES6), it is still inextricably tied to the DOM (and thus HTML elements) unless you use an abstraction like React.

It's the difference between writing something like:

```

We at MUI have been working on an open-source drag-and-drop React app builder. Link to the landing page: https://mui.com/toolpad/ This week we have published an interactive demo as well. You can check out the repo here.

Note that this uses simple OpenAI calls — you can replace this with Langchain, LlamaIndex, Hamilton (or something else) if you prefer more abstraction, and delegate to whatever LLM you like to use. And, you should probably use something a little more concrete (E.G. instructor) to guarantee output shape.

In this post, we’ll show how your team can turn any utility function(s) into reusable IPython Jupyter magics for a better notebook experience. As an example, we’ll use Hamilton, my open source library, to motivate the creation of a magic that facilitates better development ergonomics for using it. You needn’t know what Hamilton is to understand this post.

We built an app with it -- https://blog.dagworks.io/p/building-a-lightweight-experiment. You can see the code here https://github.com/DAGWorks-Inc/hamilton/blob/main/hamilton/....

Usually we've been prototyping with streamlit, but found that at times to be clunky. FastUI still has rough edges, but we made it work for our lightweight app.

This library is old news? Is there anything new that they've added that's noteworthy to take it for another spin?

[disclaimer I'm a maintainer of Hamilton] Otherwise FYI Prophet gels well with https://github.com/DAGWorks-Inc/hamilton for setting up your features and dataset for fitting & prediction[/disclaimer].

I had been hearing these pains from Langchain users for quite a while. Suffice to say I think:

1. too many layers of OO abstractions are a liability in production contexts. I'm biased, but a more functional approach is a better way to model what's going on. It's easier to test, wrap a function with concerns, and therefore reason about.

2. as fast as the field is moving, the layers of abstractions actually hurt your ability to customize without really diving into the details of the framework, or requiring you to step outside it -- in which case, why use it?

Otherwise I definitely love the small amount of code you need to write to get an LLM application up with Langchain. However you read code more often than you write it, in which case this brevity is a trade-off. Would you prefer to reduce your time debugging a production outage? or building the application? There's no right answer, other than "it depends".

To that end - we've come up with a post showing how one might use Hamilton (https://github.com/dagWorks-Inc/hamilton) to easily create a workflow to ingest data into a vector database that I think has a great production story. https://open.substack.com/pub/dagworks/p/building-a-maintain...

Note: Hamilton can cover your MLOps as well as LLMOps needs; you'll invariably be connecting LLM applications with traditional data/ML pipelines because LLMs don't solve everything -- but that's a post for another day.

This is me. I drive an open source library Hamilton that people doing time-series/ML work love to use. I'm building a paid product around it at DAGWorks, and I'm after feedback on our current version. Can I entice anyone to:

From a nuts and bolts perspective, I've been thinking of building some reactivity on top of https://github.com/dagworks-inc/hamilton (author here) that could get at this. (If you have a use case that could be documented, I'd appreciate it.)

Most people don't track lineage because it's difficult (though if you use something like https://github.com/DAGWorks-Inc/hamilton to write your pipeline - author here - it can come almost for free).