🌟 GrowEasy – AI‑Powered Inventory & Pricing Optimization for Shopify Unlock smart retail management with real-time demand forecasting, automated insights, and a full serverless architecture—built entirely on AWS with SageMaker, Lambda, DynamoDB, and React.

Shopify 🛍️ is a popular e-commerce platform, Shopify is selected as the primary integration platform for this project.

• Shopify (primary via webhooks & Admin API), WooCommerce (REST API), BigCommerce (official APIs), Magento / Adobe Commerce (enterprise-scale forecasting).

This repository contains backend webhook ingestion, ML forecasting pipeline, and a React dashboard — built during rapid development, so structure reflects functional grouping rather than strict layering.

GrowEasy/ │ ├── Backend/ # Core backend logic (Shopify → AWS pipeline) │ │ │ ├── Register_Webhook/ │ │ ├── register_webhook.py # Registers Shopify webhooks via Admin API │ │ └── GrowEasy.txt # Notes / webhook payload reference │ │ │ ├── Webhook_Handler/ │ │ ├── webhookhandler.py # Processes incoming webhook POST requests │ │ └── what it does.txt # Explanation of handler logic │ │ │ ├── forecast_result_writter/ │ │ ├── forecastResult.py # Writes ML forecast output │ │ └── what it does.txt # Description of forecasting output logic │ │ │ ├── forecast_trigger/ │ │ ├── index.js # Triggers forecasting pipeline (EventBridge/Lambda) │ │ └── what it does.txt # Notes on trigger behavior │ ├── data/ # Dataset used for ML pipeline │ ├── befor-forcast.csv # Raw historical product/order data │ └── after forecast.txt # Forecast output results │ ├── Sagemaker/ # ML experimentation & training │ └── forecasting-notebook.ipynb # LSTM model training notebook │ ├── codex 2.0/groweasy/ # Frontend (hackathon dashboard build) │ │ │ ├── public/ # Static assets │ ├── src/ # React components & UI logic │ ├── package.json │ └── tailwind.config.js │ ├── assets/ # Supporting files (images, references) │ ├── README.md # Project documentation ├── LICENSE # License file ├── Dashboard.png # UI preview ├── api_gateway_event.png # API flow proof ├── cloudwatch_logs.png # Monitoring proof └── dynamodb_tables.png # Data storage proof 

GrowEasy integrates directly with Shopify using webhooks and a serverless backend pipeline to enable real-time data processing.

App Setup & Installation
The GrowEasy app is created via the Shopify Partner Dashboard and installed on a merchant’s store using OAuth, granting secure access to product and inventory data. Webhook Registration Once installed, the system registers webhook events (e.g., product create/update) using the Shopify Admin API to receive real-time store activity. Event Ingestion (API Layer)
Shopify sends webhook POST requests to an API Gateway endpoint, which acts as a secure entry point for all incoming data. Processing (Lambda Backend) AWS Lambda processes the incoming payload, extracts relevant fields (product, inventory, pricing), and prepares it for downstream usage. Data Storage
Processed data is stored in DynamoDB for further analysis, forecasting, and historical tracking.

📷 Diagram: Webhook Flow

GrowEasy runs an automated, serverless ML pipeline to generate demand forecasts:

• ⏰ Event Trigger (EventBridge) → Schedules forecasting job at regular intervals
• 📁 Data Extraction (Lambda) → Reads historical product data from DynamoDB and exports to S3 (products.csv)
• 🧠 Model Execution (SageMaker / Python) → Trains and runs LSTM-based forecasting on product demand
• 📊 Result Storage (DynamoDB) → Stores forecast outputs per product for fast retrieval
• 📤 Alerting (SNS) → Sends notifications on low stock or demand spikes

⚡ Outcome: Automated, repeatable forecasting pipeline with no manual intervention

The system is monitored end-to-end for reliability, visibility, and debugging:

• 🔗 API Layer (API Gateway + Lambda) → Serves latest forecast data to the dashboard
• 📊 Visualization (React + Recharts) → Displays trends, forecasts, and usage insights
• 🔍 Tracing (X-Ray) → Provides request-level visibility across services
• 🔐 Access Control (IAM) → Enforces secure, least-privilege access

⚡ Outcome: Full visibility into data flow, system health, and model execution

Why this design?

• Serverless → no infra overhead
• API abstraction → secure frontend
• Event + request hybrid → supports both real-time + scheduled insights

📷 Diagram: Monitoring & Logging via CloudWatch

📷 Screenshot: GrowEasy Dashboard UI This dashboard is built specifically for retail business owners and inventory managers to monitor forecasted demand, stock levels, and price recommendations in real-time. But, Now here , Dashboard fully developed and owned by me as part of the GrowEasy AI-powered retail platform. Responsible for architecture, design, data integration, and deployment. 🔗 Live Dashboard Demo — You can explore the working example here.

☁️ Cloud & Backend

• AWS Lambda — serverless processing for webhooks, data pipelines, and forecasting
• API Gateway — secure endpoint for Shopify webhook ingestion
• DynamoDB — scalable storage for product, order, and forecast data
• Amazon S3 — stores exported datasets for ML processing

🔗 Integration

• Shopify Admin API — access product, inventory, and order data
• Shopify Webhooks — real-time event ingestion (products, updates, orders)

🧠 Machine Learning

• Amazon SageMaker — LSTM-based demand forecasting

⚙️ Orchestration & Automation

• Amazon EventBridge — schedules periodic ML jobs and pipeline execution

📊 Monitoring & Alerts

• Amazon CloudWatch — logs, metrics, and system monitoring
• Amazon SNS — notifications for low stock and demand spikes

💻 Frontend

• React + Tailwind CSS — responsive dashboard UI
• Recharts — data visualization (trends, forecasts, insights)

🔐 Security

• AWS IAM — role-based access control with least privilege

1️⃣ Clone the repository

git clone https://github.com/your-username/groweasy.git

2️⃣ Setup Backend

• cd GrowEasy/Backend
• Update webhook endpoint (API Gateway URL)
• Deploy Lambda functions (AWS Console / ZIP)

3️⃣ Configure Shopify Webhooks

• Add API Gateway endpoint in Shopify
• Subscribe to:

products/create products/update orders/create 

4️⃣ Run Forecasting Pipeline

• Upload historical data to /data
• Run: Sagemaker/forecasting-notebook.ipynb

5️⃣ Run Frontend Dashboard

cd "codex 2.0/groweasy" npm install npm start 

I built GrowEasy during the Codex 2.0 Hackathon in just 36 hours, working under extreme pressure with my team.

• ✅ Real-life problem solving → tackled challenges faced by small Shopify merchants like stockouts and wrong pricing.
• ✅ Hands-on project building → designed and implemented the end-to-end AI + cloud pipeline (Shopify → AWS → AI → Dashboard).

Here are some snapshots from my Codex 2.0 journey while building GrowEasy:

Also, you can check from here: LinkedIn

✨ This experience made GrowEasy not just a project, but a real proof of my skills in cloud, AI, and problem-solving under pressure.

Failure Handling

• Retry-safe execution (idempotent Lambda design)
• Partial failures don’t break full pipeline
• Errors logged in CloudWatch for traceability

Security

• IAM roles with least privilege (no over-permissioned access)
• No secrets exposed to frontend
• API Gateway enforces controlled access + CORS

Scalability & Performance

• Lambda auto-scales with request load
• DynamoDB handles high-throughput reads/writes
• Stateless design avoids bottlenecks

Key Decisions

• Serverless over containers → faster setup, lower ops cost
• API abstraction over direct AWS calls → security + simplicity
• Real data over mock dashboards → credibility
• Modular design → Lambda, API, frontend decoupled

Trade-offs

• Limited control over long-running workloads
• Forecasting kept lightweight (not full ML pipeline infra)
• Depends on AWS API availability

Explicit Limitations

• No enterprise-grade RBAC/auth system
• Forecasting accuracy depends on historical data quality
• Not a full FinOps platform (focused scope)

GrowEasy is an open-source initiative, and we welcome contributions from developers, data scientists, cloud engineers, and e-commerce enthusiasts!

1. 🍴 Fork the repo
2. 📦 Create a new feature branch: git checkout -b feature-name
3. ✅ Make your changes and test them
4. 📬 Submit a pull request describing your enhancement

🤝 Let's Build This Together! Made with 💚 by Manas Gantait