A framework to make robotics accessible and easy for everyone (ML engineers/firmware engineers/vibe coder/people with no technical knowledge) where all the stacks have clear separation so every speciality can focus on what they do best.
The point is that bad code pretty directly leads to a product that doesn't work. It might work today (though... I wouldn't bet my life on it with how hit and miss LLM code is). But a year or two from now, with people just piling on more and more poorly written code, the system is going to suffer. It'll be slow, it'll be buggy, etc. Yeah, your users won't be able to say "aha, this is because they used AI!", but they will certainly notice the negative effects from you having done so.
Why do you assume the code will be bad in the first place?
Coding as someone without experience in coding? Most probably yes, but from someone with some kind of expertise who can act as a guardrail for bad code piling up? Probably not.
I don't assume the code will be bad, I directly observe on a daily basis that it's bad. Since the widespread adoption of AI, all but the best developers I work with have been writing worse code with a higher number of more severe bugs.
The AI spits out _some_ code => the developer needs to review & cleanup the mess, or accept that it worsen the system for the sake of delivering more rapidly.
That's the dynamic with AI and a lot of developers hate working like that, yet there is pressure to work this way.
For me it is hard to invest too much time into being in first crowd.
This space is moving too fast for me and I have current job to do that is paying my bills.
I can invest time to watch/follow people from the first crowd.
But no one is going to give you a medal and it is not a „better crowd”.
I might need to pay for this expertise some day, but I guess it will be OpenAI or Anthropic that takes my money just like so far all the advances were introduced to frontier models or their own tooling.
https://peppy.bot a robotics framework to replace ROS 2, currently working on automatic nodes resolution so pulling a working software for your robot is as easy as doing an `apt get install my_robot`.
The goal is to enable devs from all fields (embedded devs/AI engineers/vibe coders) to work on their own stack without ever having to worry about the other parts of the robot stack.
Hey,
too bad you've shut down, looked like a great project.
> I notice you don't have shared memory transport, nor do you support runtime composability (I think?)
We're using Zenoh in the backend, we didn't try to reinvent the wheel on that front.
> I'm also a little confused on what serialization format you support - is it an entirely custom one?
It's cap'n proto, but abstracted away for the user so he doesn't have to think about it.
> BSL feels great, but I found it scared off some people
Agreed, but we're building a SaaS in parallel, PeppyOS will be 100% free and open source but we don't want another company to build a SaaS on top of it, that will be the only restriction of the BSL license, otherwise it's the same as Apache 2.0.
