zestones

zestones / Aria

Public

In every factory there's one person who knows when a machine is about to fail — they hear it. When they retire, that knowledge disappears forever. ARIA captures it, watches the equipment, and diagnoses what goes wrong — so the one who knows is never the last.

aria.vgtray.fr ai anthropic factory-automation hackathon industrial
18
5
100% credibility
Found May 01, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

ARIA is an AI agent system that turns factory manuals, operator notes, and live sensor data into automatic anomaly detection, root-cause diagnosis, and repair work orders.

How It Works

1
🚀 Launch your factory dashboard

Download and start ARIA on your computer to see your machines' live status with one simple command.

2
📄 Upload your machine manual

Drop in the PDF guide for one of your factory machines so ARIA can learn its specs and limits.

3
💬 Share your expert know-how

Answer a few quick questions about normal sounds, past fixes, and quirks you notice on the floor.

4
👀 Watch machines in real time

Sit back as ARIA monitors vibrations, temperatures, and flows from your equipment 24/7.

5
🚨 Catch problems early

Get instant alerts with smart diagnosis and ready-to-print repair plans when something drifts wrong.

6
🗣️ Chat with your assistant

Ask natural questions about issues, stats, or history, and get clear answers with charts.

✅ Factory runs smoother

Save hours spotting failures early, with less downtime and knowledge that sticks around forever.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 18 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is Aria?

Aria captures retiring operators' intuition about factory machines failing—vibration hums, log notes, handover whispers—before it's lost forever. Drop in a PDF manual, answer four questions, and it builds a knowledge base, ingests live signals from simulators or real sensors, then deploys Claude-powered agents to detect drifts, diagnose root causes, and spit out work orders. Built on Python FastAPI backend with React/TypeScript frontend and TimescaleDB, it spins up the full stack via Docker Compose and three Makefile commands, complete with seeded demo data for every factory scenario.

Why is it gaining traction?

Unlike €500k enterprise CMMS setups taking six months, Aria goes live in 10 minutes with PDF vision extraction and operator calibration—no specialists needed. Hackathon winner for best Claude managed agents, it demos predictive alerts and sandboxed diagnostics out of the box, blending realtime signals, KPIs like OEE/MTBF, and agent handoffs in a hot-reload dev loop. Devs dig the architecture docs covering everything from MCP tools to simulators, making it a blueprint for AI in arknights endfield every factory or satisfactory builds.

Who should use this?

Maintenance engineers prototyping predictive systems for pumps, fillers, or UV sterilizers in bottling plants. IoT devs evaluating Claude agents for signal anomaly detection without wiring PLCs. Hackers exploring TimescaleDB hypertables for every factory you need in satisfactory, skipping arduino nano every github tweaks.

Verdict

Grab it for learning agentic workflows—docs are gold, Makefile shines, CI solid—but 18 stars and 1.0% credibility scream early prototype, not prod-ready. Fork for your aria ai github experiments; ignore distractions like ariana grande parfum or ariane 6.

(187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.