If you like chat interfaces, you might like this open-source tool I built, which is basically a chat UI but it saves you the hassle of copying and pasting files https://github.com/dyad-sh/dyad
Basically it's running a chat UI locally on a Git repo and you can reference files like "#foo.py" and then if you want to edit a file, you hit "Apply code" for a markdown code block and then it shows you a code diff so you can review the changes before actually updating the files. I've found this makes it much easier to curb stray edits and keep the LLM edits focused on what you actually care about.
If you're just getting started and want to keep using Pycharm, you can check out https://dyad.sh/ it's an open-source standalone AI pair programming tool so you don't need to switch IDEs. It's also focused on making it like you're pair programming with a real engineer, asking you clarifying questions and having you pick from different options when the request is unclear.
HN has a strange philosophy built into its ranking algorithm that an item with a large number of comments early on should be de-ranked because the conversation is likely to be of poor quality.
Hey folks - I shared about Mesop on Show HN (https://news.ycombinator.com/item?id=40567327) ~2 months ago and got a ton of really helpful feedback. For those unfamiliar, Mesop is a new open-source Python UI framework that enables you to rapidly build AI apps.
We often get asked how Mesop differs from other Python UI frameworks like Streamlit and Gradio so we've created a comparison guide between Mesop and Streamlit and Gradio.
Hey folks - I shared about Mesop on Show HN (https://news.ycombinator.com/item?id=40567327) ~2 months ago and got a ton of really helpful feedback. For those unfamiliar, Mesop is a new open-source Python UI framework that enables you to rapidly build AI apps.
We often get asked how Mesop differs from other Python UI frameworks like Streamlit and Gradio so we've created a comparison guide between Mesop and Streamlit and Gradio.
Any particular parts that you feel are inadequate with web components? Web components aren't a panacea but they have definitely come a long way in the past decade.
Thanks, yeah describing Mesop API as essentially a UI over RPC is a good analogy and is basically what's happening under the hood [1].
I guess one funny thing is that I actually identify as a FE developer more than a Python developer :) I've been doing FE for much longer than Python and agree that there's warts on both sides (and no language/ecosystem is perfect). For me, a lot of the FE fatigue comes from all the inherent challenges with developing client-side applications (e.g. the necessity of bundling/minifying for optimizing performance, transpiling for legacy browsers) so I don't say this to throw stones at the other side but simply to point out my own fatigue having dealt with this complexity for years.
Right now Mesop is definitely more focused on the demo use cases, but we're also trying to push the limits of the kinds of apps you can build in Python so it doesn't have to be something you throwaway when you want to deploy a scalable app.
Since then, we've made a lot of updates to Mesop, including adding support for web components which allows you to write custom JS and wrap existing JS libraries.
We've also fully re-designed our home page at https://google.github.io/mesop
- with many more code examples and demo, which hopefully shows how Mesop is different than some of the other Python UI frameworks out there.
As always, appreciate any feedback about the project. Thanks!
You've also pivoted a bit into AI? :) Personally I find that unfortunate because that's not what you actually do (even though many people at Google might use the framework for this) and it's distracting.
Thanks! I've used Google's internal version of Bazel but it wasn't until this project where I had to spend a lot of effort getting Bazel working in a new project, which is honestly a lot of work and not very straightforward :(
What helped me the most was looking at other projects using Bazel with similar tech stacks and then assembling it together, e.g. :
Basically it's running a chat UI locally on a Git repo and you can reference files like "#foo.py" and then if you want to edit a file, you hit "Apply code" for a markdown code block and then it shows you a code diff so you can review the changes before actually updating the files. I've found this makes it much easier to curb stray edits and keep the LLM edits focused on what you actually care about.
reply