Renvy Python Library Contribution

[sharp wharf]

Description Of The Library

Renvy is a python animation library that is written in rust. It focuses on simplicity, performance and flexibility, the library is a similar one to what you find in https://www.manim.community or https://motioncanvas.io but made for general use-cases. It features:

• 2D/3D Rendering It will support both 2D and 3D rendering scenes, although 3D scenes for now will be limited
• Presentation Mode When lets say you do want to use the library for presenting something, u could enable presentation mode to pause the animation and then continue on
• Various Shapes It will include various shapes, both common, special and flexible ones
• Flexible Data Strutures things like colors, vectors... etc.
• Tweens this one is an especially vital part of the library. Consider viewing this diagram to show how its processed. They are heavily-inspired by how godot handles them thus why it looks godot-like
• Realtime Preview. You can preview a scene without exporting

Who Am I?

Im a rust intermediate fyi. BUT don't worry about that, this is mostly syntax & some specific guidelines i have to follow which il get used to overtime. If the rendering is in low-level, i do want to try and learn modern gpu programming with the help of someone who knows it well. Not nesscarrily be a teacher but help out on the hard parts while i observe how things work and also work / lead on the project

What Im Looking For Potential Contributor

• A person well mannered / kind person, not a bad apple
• A person knows the ins and outs of rust(not perfectly but still very good enough)
• A person that knows low-level rendering (if rendering will use wgpu)

A Previous Similar Project

I worked on a minecraft mod using java with a partner of mine(they helped on some nuisances while i was thinking more deeply about the library) https://github.com/GitBrincie212/Apel-Mod/tree/unstable-version-work

[sharp wharf]

hit me up with a DM if ur interested

[valid gazelle]

Does your project internals use rust? No just curious 😅 because you mention that it is a python library so I am quite confused like does your library use rust or not

[sharp wharf]

it uses rust internally for the reason of performance

also to ensure some type safety

but

it uses pyo3 for sort of "transcode" it to a python library

[valid gazelle]

Ah ok I see, because your description does not clearly mention about that. So that's why the confusion. I think I would suggest maybe you can update the description a little bit to make it more clear so that no one else confused. Other than that cool 🚀 project btw 🙂

[sharp wharf]

thx

il revamp the description a bit

i've changed the upper part

is it more clear?

[valid gazelle]

Also, is it a paid project or something? I am sorry for the confusion again but I don't see any project link so that's I am asking this. 🙂

[sharp wharf]

nope

completely passion-driven

and yeh the reason u don't see is because well i haven't really made it. Its the beggining part of the project

[valid gazelle]

Yes ok it is more clear now

[sharp wharf]

tbh i probably will use my library for videos

but apart from that i do want it open-source, free to use for anyone and be a great tool

im also the only person on the team, if u can call it a team for now

[valid gazelle]

Ah ok I see, this is like a more of a proposal of your idea cool cool. I think I can help suggest you some crates for this:

• For gpu processing: rust-gpu can help but be aware that it requires unsafe for now.
• For video processing I think the rust-cv can help you with it.
• I am not sure but if you like want to completely automate everything like the subtitles and speech I think there is crate called ollama which can help you with that.

I hope this helps 🙂

[sharp wharf]

hmm noice

tbh rn want to display things]

like in a window

tbh wish i had some partner to work with this on. I really want to turn this to a reality

its difficult doing stuff alone, even on other languages

[valid gazelle]

Also, for inspiration I think I would also suggest this video too which can help frame the idea of the project more strongly:

https://www.youtube.com/watch?v=_rGXIXyNqpk

Though it is in other languages but gives the idea. 😅

[sharp wharf]

AI won't be integrated to this

its not a great use-case for a library

there are certain advantages to write code rather than using something like adobe after affects, final cut pro x

Its more modular. It allows for defining systems that can be used extensively

wanna animate this?

good luck buddy as there are a lot of lines

[valid gazelle]

Yes true agreed. Code is more flexible than AI. Also, it can act as a reusable template too maybe 🙂

[sharp wharf]

yeh true

AI has a lot of hype around it tbh, both being good as its quite a revolutionary thing i do agree. But this hype blinds us into incorporating ai everywhere and anywhere

[valid gazelle]

Ah so basically simple and complex animations with just a few lines of code. Ok I see the potential. This can be used in variety of places like I am already see the use cases now. 😅

[sharp wharf]

yeh

also the animation is from manim

just to illustrate an example

also, the project used to be like this

this is the second iteration of the project

it represents the different easing curves

[sharp wharf]

if ur interested tbh, u can try to sort of lend a hand for helping

we can build this if we both got patience and a good vision(not overscoping things)

[sharp wharf]

ig nobody

interested

[sharp wharf]

its quite a hard project to tackle due to the somewhat limited ecosystem in terms of graphics

[sharp wharf]

its either having to deal with overly complex low-level graphics

or

having too many systems

[sharp wharf]

imma strip away for now the 3D scene feature

the reason being is that its complex to make it

imma use speedy2D

il probably need to refactor the code in order to support 3D or extend speedy2D ik

but for now imma focus on velocity

[sharp wharf]

yeh tbh im considering to drop it for now.

1. There isn't enough interest
2. idrk about low-level GPU programming so im limited on options i can take

i might try making a rendering engine in rust using wgpu

maybe it could be used in the renvy project

[last bridge]

check this out: https://github.com/yongkyuns/noon?tab=readme-ov-file. consider forking it.

[sharp wharf]

i might

damn i forgot about this project

im bombarded with other projects

tbh im gonna mostly look at how they handle graphics and all that jazz

maybe i will use nannou

[sharp wharf]

its gonna be a while before i come back to this

[faint scroll]

Bummer, was just gonna ask if you're still open to contributions/collab

[sharp wharf]

tbh im open tho i need experienced rustaceans

with a full portfolio on their work

[faint scroll]

Dig it. As soon as my project is public ready, I think you'll find I've got a few chops. I'm not the oldest Rustacean, but still. Fair ask, I'll shoot you my GitHub when it's ready, maybe a week or two

[sharp wharf]

well i don't care for way too experienced rustaceans

i just require experience

im neither a experienced rust developer

the problem is that i got a lot of projects

[faint scroll]

I get having a lot of irons in the fire. I’m currently solo-developing a modular AI runtime in Rust that implements a Zero Trust enforcement layer. It acts as a bridge between a structured canonical database and any connected model, validating outputs in real time with cryptographic signatures and reasoning chains. It’s fully database-agnostic (SQL/NoSQL), and I’m building a companion system: a signed, community-vetted code store—think Stack Overflow meets GitHub, but powered by earned trust and inverse reinforcement learning.

The goal? Auditable AI with explainable behavior. Not just plausible answers—provable ones. It’s a reinforced-by-design architecture where integrity of logic and source is prioritized over surface-level content.

The system’s wired for trust enforcement at every layer. I’ve got the schema locked, most core modules built and crated, and I’m prepping to wire everything into an Alpine Linux base. Once that’s done, I’ll Dockerize the whole thing and either run it against a local model (Phi-4 + Reasoning for early trust testing), or connect via API to a custom GPT deployment. Either way, it’s a zero trust sandbox—ready to validate every output, every time.

I've been pretty quiet up until now because I haven't been close enough to validate proof of concept until recently.

[sharp wharf]

ok

im not an ai guru guy

[last bridge]

@faint scroll what's a "Zero Trust enforcement layer" - specifically in what way is it zero trust?

[faint scroll]

@last bridge Great question, let's tackle it. “Zero Trust” in the Sigil Protocol means that no model output, nor human input, is inherently trusted—ever.

Just like in cybersecurity, where Zero Trust means no user or device is trusted by default, Sigil applies the same principle to cognition:

Every action, suggestion, or output from a model must justify itself. Every access to Canonical data is logged, traced, and validated. Every trust decision flows through a verifiable chain of reasoning.

This is enforced by:

A reasoning chain (Input → Context → Reasoning → Suggestion → Verdict → Audit → IRL Score), cryptographically signed Canon and model checkpoints, a real-time Inverse Reinforcement Learning (IRL) engine that evolves based on what trusted behavior looks like over time, and LOA-bound access controls that limit model behavior based on verified credentials.

TL;DR: If a model says something, Sigil asks, “Why should I believe you?”
If it can’t explain itself or show its receipts, it doesn’t pass.

That’s Zero Trust—but applied to cognition instead of networks.

The work isn't centered around model choice, either. I geared it to be model-agnostic, so I could either use a local model like Phi4+Reasoning or slap a Claude/ChatGPT/Grok/etc API key in and watch it go. Because of the IRL engine and LOA scoped access, it really doesn't matter what model you plug in; it's essentially a secure operating system, but for AI/Human interaction rather than human/pc style.

[sharp wharf]

💀

that ain't the place to talk here

[swift marlin]

superuser666_30897 has been warned, this is their first warning.

[faint scroll]

My bad, dude asked. No disrespect meant

[cunning dagger]

I'll leave the messages up since someone did interact with them, but the original message was an obvious derail.

[faint scroll]

I waas empathizing with McBrincie. Dude's working on a lot; dig it. Should've goven a TL;DR. Noted

[sharp wharf]

well...

i think i can sort of work

on the project

[sharp wharf]

im thinking of making this bigger and support multiple programming languages (Rust, Python, Javascript/Typescript)