Skip to content

updated readme #55

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
idky137 merged 1 commit into from
Apr 14, 2026
Merged

updated readme #55

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
24 changes: 16 additions & 8 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,22 +6,30 @@
[![License](https://img.shields.io/badge/license-Apache--2.0-blue.svg)](LICENSE-APACHE)
[![MSRV](https://img.shields.io/badge/MSRV-1.81-blue.svg)](#)

> **Portable, contract-enforcing computation graphs for AI-enabled embedded
> systems — from bare-metal microcontrollers to multi-threaded servers, in
> safe Rust.**
> **A real-time runtime for continuous AI systems operating on live data
> streams — portable, contract-enforcing computation graphs from bare-metal
> microcontrollers to multi-threaded servers, in safe Rust.**

> **Alpha release.** Core contracts, code generator, and integration-tested
> runtimes are functional. APIs may change before v0.1.0.

> See [umbriel-systems.com/limen](https://umbriel-systems.com/limen) for
> project overview and target domains.

---

## The Problem

Building AI-enabled robotics and edge computing systems today means
rebuilding the same software pipeline for every hardware target. A
perception-inference-actuation pipeline written for a Cortex-M4 cannot run
on a Raspberry Pi or a server without substantial rework — despite
implementing identical logic.
AI systems are increasingly deployed in environments where data is
continuous, latency constraints are strict, and resources are limited.
Robotics platforms, industrial control systems, and edge sensor networks
require real-time processing of live data — not batch jobs or API calls.

Existing infrastructure — from PyTorch to ONNX Runtime — is designed for
batch or request-response workloads. These systems introduce unpredictable
latency, dynamic memory behaviour, and limited control over execution. And
building for a Cortex-M4 means rebuilding everything when you move to a
Raspberry Pi or a server — despite implementing identical logic.

No existing framework bridges this gap:

Expand Down