LODEXI Core Engine

High-Performance, Multi-Tenant Knowledge Indexing & Grounded Retrieval Engine

Overview

LODEXI Core is the intelligence backend of the LODEXI ecosystem. It is a decoupled, ultra-fast RAG (Retrieval-Augmented Generation) engine built with Python, FastAPI, and Qdrant Vector Database.

It provides semantic search, document ingestion, and grounded conversational QA capabilities for the lodexi-portal and any other external client applications.

---

Key Features

1. Strict Multi-Tenant Isolation: Enforces tenant boundary partitioning via X-API-Key validation and Qdrant payload filtering. Zero cross-tenant data leakage.
2. Dual-Mode Serving:
   • POST /v1/search: Pure semantic vector retrieval returning ranked chunks, scores, and metadata without LLM synthesis.
   • POST /v1/ask: Grounded conversational synthesis providing factual answers with verifiable citations and source links.
3. Decoupled & Language-Agnostic: Easily connect any web app (Next.js, Laravel, Go, Node.js) via simple JSON REST requests.
4. Visual Dashboard: Includes a built-in dashboard.html visualizer matching the LODEXI branding aesthetics.
5. Flexible Runtime: Runs 100% locally with embedded Qdrant (zero Docker required) or scaled via Docker Compose.

---

Project Architecture

lodexi-core/
├── requirements.txt            # Python dependencies
├── .env.example                # Configuration template
├── src/
│   ├── main.py                 # FastAPI application entrypoint
│   ├── config.py               # Pydantic environment settings
│   ├── core/                   # Security, Vector Store, LLM configs
│   ├── models/                 # Request/Response Pydantic schemas
│   ├── templates/              # Visual UI Dashboard
│   └── api/                    # v1 REST Endpoints
└── tests/                      # Automated Pytest suite

---

🛠️ Quickstart Guide

1. Setup Environment

git clone https://github.com/lodexi/core.git lodexi-core
cd lodexi-core

# Create and activate virtual environment
python -m venv .venv

# Windows:
.\.venv\Scripts\activate
# Mac/Linux:
# source .venv/bin/activate

# Install dependencies
pip install -r requirements.txt

# Copy environment settings
cp .env.example .env

2. Run the Engine

uvicorn src.main:app --reload --port 8000

• Visual Dashboard: http://localhost:8000
• Swagger API Docs: http://localhost:8000/docs

3. Run Automated Tests

pytest

---

API Integration Examples

Example: LODEXI Portal (PHP/Laravel)

// Call LODEXI Core Grounded QA
$response = Http::withHeaders([
    'X-API-Key' => 'key_portal_secret_123',
])->post('http://localhost:8000/v1/ask', [
    'question' => 'How to setup multi-tenancy?',
    'limit' => 4,
]);

$answer = $response->json()['answer'];
$citations = $response->json()['citations'];

---

License

Copyright © 2026 LODEXI. All rights reserved.

This software is proprietary. You may not use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the software without explicit written permission.