Skip to main content
Back to Build with GLM 5.1 · Z.ai Builder Series Gallery

Project spotlight

Approved

Vyuha AI

by Null

Autonomous Multi-Cloud Recovery Orchestrator powered by GLM-5.1.

Vyuha AI

Demo Video

About This Project

Most cloud monitoring tools tell you what broke. Vyuha AI tells you what to do about it ,and then fixes it for you.

Vyuha AI is a Triple-Cloud Mission Control system that monitors AWS, Azure, and GCP nodes simultaneously. The moment a node fails ,whether it's a hard crash, intermittent packet drop (Flaky), or a massive latency spike ,the Z.ai Companion (powered by GLM-5.1) analyzes the severity and generates a dynamic failover plan.

Key Features: • Agentic AI Triage: Generates step-by-step reasoning for why traffic should be rerouted. • Human-in-the-Loop: No action is taken without operator approval. It proposes, you approve, it acts. • Evolutionary Memory: Every human decision is stored as a lesson. The AI reads its own history before every new triage, getting smarter with every incident. • Zero-Downtime Proxy: Dynamically isolates dead data centers and reroutes user traffic instantly. • Real-time Chaos Lab: Inject Hard Kills, Flaky nodes, and Latency spikes live to test the system's resilience.

Built for the Z.ai Hackathon to prove that cloud infrastructure should be as intelligent as the team running it.

Repository

Vyuha AI: Autonomous multi-agent orchestration for real-world traffic and chaos, powered by GLM 5.1.

Python65.1%TypeScript32.2%Dockerfile1.5%CSS0.8%JavaScript0.3%
Last commit 6 months ago