ketan_around

Yeah, haha funnily enough I'm a huge fan of both C4 and plantuml. We actually do have an experimental 'software map' feature, turned off in settings by default, which draws a hierarchical / C4-based architecture diagram of your codebase (even highlighting the nodes which changed in the PR!) Warning, it is a little rough around the edges at the moment.

As far as plantuml/graphviz/mermaid/similar UML tools, it's really cool stuff, I agree. My issue is that those tools are mainly built to render static diagrams (output to svg, png, etc.) Some eg mermaid are perhaps extensible for interactivity, but it's a little wonky. I think a lot of the value of whiteboard comes from the interactivity -- being able to click on a node to see the underlying code, being able to go through a guided walkthrough of a given flow, etc.

But using their UML authoring format is certainly worth looking into, since the LLMs are already so well-versed in it. i'll put some thought into this!

yeah, for sure. i think the diagrams/artifacts are definitely more ephemeral, since they fall out of date. in my experience, the more lasting artifacts from the whiteboarding session have been 1) making the code that ships higher quality because i understand it, and 2) getting my own mental model aligned with the current state of the codebase.

we're dreaming up some features in a hosted version of the product which could maintain an up-to-date 'software map' of your codebase, and that is the backbone which would power all diagrams/visualizations

for sure, we're thinking of doing something along those lines for our hosted product-- auto-trigger reviews upon opening PR, link to them from gh, etc. feel free to shoot me a message (ketan at dev dot fast) if you're interested in that!

not no-code at all (a big part of this project has been our semantic diffing engine)! check out the demo - in fact if anything it's meant you bring more into touch with the code, rather than less. :)

free, oss, and local-only right now! we are working on a hosted solution but honestly aren't sure yet; we mostly made this for ourselves to fix our own gripes with agentic coding :)

Emacs was the first vibe-coding.

Exciting to see a product like this launch! There are obviously a host of ‘memory’ solutions out there that try to integrate in fancy ways to cache knowledge / save tokens, but I think there’s a beauty in simplicity to just having a proxy over the OpenAI endpoint.

Interested to see where this goes!