IntelliHybrid is a production-ready, AI-powered framework that enables secure, seamless bidirectional communication between on-premise infrastructure and AWS cloud — with natural language querying, auto-generated schema documentation, and zero code changes after setup.
IntelliHybrid bridges the gap between your on-premise data center and AWS cloud by:
- 🔐 Establishing secure VPN tunnels (Site-to-Site, OpenVPN, or Direct Connect)
- 🗄️ Connecting on-prem databases (MySQL, PostgreSQL, Oracle, SQL Server) to AWS
- ⚡ Auto-provisioning DynamoDB tables with your custom Partition Key (PK) and Sort Key (SK)
- 🔄 Bidirectional data synchronization — on-prem → cloud and cloud → on-prem
- 🛡️ Enterprise-grade security — IAM roles, KMS encryption, TLS everywhere
- 🤖 AI-powered data intelligence — query in plain English, auto-generated column descriptions, instant data dictionaries
┌─────────────────────────────────┐ ┌──────────────────────────────────┐
│ ON-PREMISE │ │ AWS CLOUD │
│ │ │ │
│ ┌─────────────┐ │◄───────►│ ┌──────────────────┐ │
│ │ Your DB │ │ VPN / │ │ DynamoDB │ │
│ │ MySQL / │ IntelliHybrid│ Direct │ │ Tables (auto- │ │
│ │ Postgres / │◄──────────────►│ Connect │ │ provisioned) │ │
│ │ Oracle │ │ │ └──────────────────┘ │
│ └─────────────┘ │ │ │
│ │ │ ┌──────────────────┐ │
│ ┌─────────────┐ │ │ │ AI Assistant │ │
│ │ App Server │ │ │ │ NL Queries + │ │
│ │ (any lang) │ │ │ │ Schema Intel │ │
│ └─────────────┘ │ │ └──────────────────┘ │
└─────────────────────────────────┘ └──────────────────────────────────┘
▲ ▲
└──────── IntelliHybrid ───────┘
config.yaml + Claude AI drives all
Experience natural language queries, schema intelligence, and data dictionaries in your browser — no setup needed.
IntelliHybrid includes a full AI layer powered by Claude that understands your DynamoDB tables. No more writing queries — just ask.
Write questions the way you'd say them out loud. IntelliHybrid translates them into the correct DynamoDB operation and returns live data.
from src.ai.assistant import AIAssistant
assistant = AIAssistant(config, anthropic_api_key="sk-ant-...")
result = await assistant.chat("Show me all orders from customer C-001")
result = await assistant.chat("How many products have stock below 10?")
result = await assistant.chat("Find users who signed up this month")
result = await assistant.chat("Add a new user: Jane Doe, email jane@example.com")Examples of questions that just work:
| What you type | What runs |
|---|---|
"Show me all orders from customer C-001" |
query with KeyConditionExpression |
"How many products are low on stock?" |
scan with FilterExpression |
"Find the 5 most recent signups" |
scan with Limit + sort |
"Get order ORD-8821" |
get_item with exact key |
"Delete expired session sess-99" |
delete_item |
The AI also maintains conversation history, so follow-up questions like "now filter those by electronics" work naturally.
IntelliHybrid reads your table schema and a few sample rows, then generates clear business-friendly descriptions for every attribute — automatically.
from src.ai.schema_intelligence import SchemaIntelligence
intel = SchemaIntelligence(config, anthropic_api_key="sk-ant-...")
description = await intel.describe_table("orders-table"){
"table_description": "Stores customer order transactions with fulfillment status tracking",
"attribute_descriptions": {
"orderId": "Unique identifier for each order transaction",
"customerId": "References the placing customer — links to users-table PK",
"status": "Fulfillment state: processing, shipped, delivered, or cancelled",
"total": "Order value in USD cents",
"createdAt": "Unix timestamp when placed, used as sort key for date-range queries"
},
"access_patterns": [
"Query all orders for a specific customer",
"Get a single order by ID",
"Filter orders by status for a fulfillment dashboard"
],
"suggestions": [
"Consider adding a GSI on status+createdAt for pipeline queries"
]
}One call produces a fully formatted markdown data dictionary — ready for wikis, compliance documentation, or technical portfolios.
dictionary = await intel.generate_data_dictionary("orders-table")
print(dictionary)# Data Dictionary: `orders-table`
**Purpose:** Stores all customer order transactions...
## Attributes
| Attribute | Type | Description |
|--------------------|------|------------------------------------------------------|
| `orderId` 🔑 PK | S | Unique identifier for each order transaction |
| `customerId` 🔑 SK | S | References the customer — links to users-table PK |
| `status` | S | Fulfillment state: processing, shipped, delivered... |
| `total` | N | Order value in USD cents |
| `createdAt` | N | Unix timestamp, used for date-range queries |
## Common Access Patterns
1. Query all orders for a specific customer
2. Get a single order by ID
3. Filter orders by statusExpose the entire AI layer over HTTP — connect any frontend, dashboard, or external tool.
pip install "intellihybrid[ai]"
export ANTHROPIC_API_KEY="sk-ant-..."
uvicorn src.ai.server:app --host 0.0.0.0 --port 8080# Ask your data a question
curl -X POST http://localhost:8080/chat \
-H "Content-Type: application/json" \
-d '{"message": "Show me all orders from customer C-001"}'
# AI schema description for a table
curl http://localhost:8080/tables/orders-table
# Full data dictionary (markdown)
curl http://localhost:8080/tables/orders-table/dictionary
# Suggested example queries
curl http://localhost:8080/tables/orders-table/suggestionsInteractive Swagger docs available at
http://localhost:8080/docs
The file docs/ai-demo.html is a fully self-contained interactive demo of the AI Assistant. It runs entirely in the browser with no backend needed.
To view it live, enable GitHub Pages:
- Go to Settings → Pages in this repo
- Set Source to
mainbranch,/docsfolder - Click Save
Your demo will be live at:
https://clever-boy.github.io/IntelliHybrid/ai-demo.html
You can also open it locally — just double-click docs/ai-demo.html in your file browser. It shows:
- 💬 Chat interface with AI query responses and record cards
- 🗂️ Sidebar with connected tables and clickable example questions
- 📋 Query result display with per-row attribute cards
- 📖 AI-generated schema description view
- 📚 Auto-generated data dictionary output
pip install intellihybrid
# or from source:
git clone https://github.com/Clever-Boy/IntelliHybrid.git
cd IntelliHybrid
pip install -e .cp config/config.template.yaml config/config.yamlaws:
region: us-east-1
account_id: "123456789012"
access_key_id: "${AWS_ACCESS_KEY_ID}"
secret_access_key: "${AWS_SECRET_ACCESS_KEY}"
onprem:
database:
type: mysql # mysql | postgres | oracle | mssql
host: 192.168.1.100
port: 3306
name: production_db
username: "${DB_USER}"
password: "${DB_PASSWORD}"
vpn:
type: site-to-site
customer_gateway_ip: "203.0.113.10"
dynamodb:
tables:
- name: orders-table
partition_key: { name: orderId, type: S }
sort_key: { name: customerId, type: S }
billing_mode: PAY_PER_REQUESTintellihybrid init --config config/config.yaml- ✅ Validates all credentials
- ✅ Establishes VPN tunnel
- ✅ Creates DynamoDB tables with your PK/SK schema
- ✅ Sets up least-privilege IAM roles
- ✅ Runs a connectivity health-check
intellihybrid sync --mode bidirectional --interval 60- All secrets via environment variables or AWS Secrets Manager — never in config files
- KMS-encrypted DynamoDB tables by default
- TLS 1.3 for all data in transit
- Least-privilege IAM roles auto-generated per table
| Database | Version | Status |
|---|---|---|
| MySQL | 5.7, 8.0+ | ✅ Full |
| PostgreSQL | 12+ | ✅ Full |
| Oracle | 19c+ | ✅ Full |
| SQL Server | 2019+ | ✅ Full |
| MongoDB | 5.0+ | 🔜 Coming |
intellihybrid sync --mode full # one-time complete sync
intellihybrid sync --mode bidirectional # continuous, recommended
intellihybrid sync --mode push # on-prem → DynamoDB only
intellihybrid sync --mode pull # DynamoDB → on-prem only
intellihybrid sync --interval 30 # custom interval (seconds)| Document | Description |
|---|---|
| 📖 How-To-Use Booklet | Complete step-by-step setup guide |
| 🤖 AI Features Guide | Full AI query engine & schema intelligence docs |
| 🔐 Security Guide | IAM, KMS, TLS hardening |
| 📦 Zenodo & Release Guide | DOI, download metrics, EB-1A |
| 🤝 Contributing | How to contribute |
Contributions are welcome! Please read CONTRIBUTING.md first.
pytest tests/ -v --cov=src --cov-report=htmlGood first issues:
| Feature | Difficulty |
|---|---|
| MongoDB connector | Medium |
| Terraform module | Medium |
| CDC real-time streaming | Hard |
| Web UI dashboard | Hard |
If you use IntelliHybrid in your research or production systems, please cite:
@software{kadam_intellihybrid_2025,
author = {Kadam, Shailesh},
title = {IntelliHybrid: Intelligent On-Premise to AWS Cloud Connector},
year = {2025},
publisher = {Zenodo},
doi = {10.5281/zenodo.19121004},
url = {https://github.com/Clever-Boy/IntelliHybrid}
}MIT License — see LICENSE for details.
Shailesh Kadam
🌐 GitHub @Clever-Boy
📍 Dallas, Texas
💼 LinkedIn
⭐ Star this repo if IntelliHybrid saves you time! ⭐
Your stars directly support open-source hybrid cloud tooling.