NimbusDB — Distributed AI-Native Vector Database

NimbusDB is a cloud-native, distributed, AI-ready vector database platform built with Go, Rust, and Next.js, deployed on Kubernetes using Helm.

Features & Architecture

• AI-Native Vector Engine: High-dimensional vector embeddings, HNSW ANN & exact similarity search with metadata filtering.
• Microservices Architecture: 10 decoupled microservices including Auth, Gateway, Control Plane, Metadata Service, Scheduler, Worker Nodes, Capacity Planner, SLA Monitor, Deployment Controller, and Dashboard.
• Kubernetes Native: Complete Helm charts with StatefulSets, Deployments, and automated Canary/Rolling update capabilities.

Documentation

All project documentation, implementation phases, benchmarks, and simulation walkthroughs are available in the ai-docs/ folder:

• Project Status & Architecture Architecture
• Simulation Walkthrough & Verification
• Verification Checklist
• Detailed Technical Documentation