An AI-augmented e-commerce CRM built for modern operations.

EcomCRM combines customer management, product catalog control, order tracking, analytics, revenue intelligence, Cloudinary media handling, and a Groq-powered assistant in one polished Django application.

EcomCRM is not a basic CRUD demo. It is a full-stack commerce operations product that was upgraded to feel production-ready across both engineering and product layers:

• Professional dashboard and UI refresh across the template layer
• Real revenue calculations powered by database records
• Groq-powered "ASK to AI" assistant grounded on app features and live CRM context
• Cloudinary-backed customer and product media storage
• Automatic media sync during Render builds for existing records
• Safe media fallback for older assets if a Cloudinary file is missing
• Render-ready deployment config with environment-based settings
• Analytics, tags, CSV import, maps, and operational workflows preserved

For recruiters, this project demonstrates product thinking, backend ownership, frontend polish, AI integration, deployment awareness, and production debugging under real constraints.

The ASK to AI page is connected to Groq and designed as a grounded assistant, not a generic chat box.

It uses:

• query feature extraction
• retrieval over platform capabilities and live CRM signals
• grounded answer generation
• conversation history stored in session

The assistant can explain:

• how revenue is calculated
• where to manage customers, products, and orders
• what pages exist in the system
• live operational context such as customers, products, order states, and revenue snapshots

The revenue page is calculated from real order and product records:

• delivered orders drive realized revenue
• pending and out-for-delivery orders contribute to pipeline value
• daily revenue is grouped by order date
• annual projection is derived from year-to-date performance

Media handling was upgraded to support long-term hosted storage:

• customer and product images upload to Cloudinary
• existing local media is synced during deployment
• malformed Cloudinary env values are sanitized safely
• older image records can still render through a local fallback route if needed

The UI was modernized without breaking backend behavior:

• redesigned dashboard
• upgraded customer, product, order, analytics, revenue, login, about, and contact pages
• improved footer and navbar
• consistent cards, spacing, typography, and mobile responsiveness

flowchart LR
    A[User] --> B[Django Views and Templates]
    B --> C[CRM Workflows]
    C --> D[(SQLite or Postgres via DATABASE_URL)]
    C --> E[Cloudinary Media Storage]
    C --> F[Groq Ask to AI]

    subgraph CRM Workflows
        G[Customers]
        H[Products]
        I[Orders]
        J[Analytics]
        K[Revenue]
        L[Tag Import]
    end

    F --> M[Feature Extraction]
    M --> N[Context Retrieval]
    N --> O[Grounded Response]

• Django 6
• Django ORM
• django-filter
• WhiteNoise
• gunicorn
• dj-database-url

• Django Templates
• HTML
• CSS
• Bootstrap 5
• Bootstrap Icons
• JavaScript
• Chart.js
• Leaflet

• SQLite for local development
• Postgres-compatible production setup through DATABASE_URL
• Cloudinary for hosted media storage

• Groq API
• Structured feature extraction and grounded generation

• Render
• render.yaml
• Procfile
• build.sh

git clone https://github.com/rudra00434/EcomCRM.git
cd EcomCRM

python -m venv .venv

Windows:

.venv\Scripts\activate

macOS / Linux:

source .venv/bin/activate

pip install -r requirements.txt

Copy the example file and update the values:

Windows:

copy .env.example .env

macOS / Linux:

cp .env.example .env

python manage.py migrate

python manage.py runserver

The app will be available at:

http://127.0.0.1:8000

Cloudinary value format:

CLOUDINARY_URL=cloudinary://API_KEY:API_SECRET@CLOUD_NAME

Important for Render:

• paste only the value
• do not include angle brackets
• do not wrap the value in extra quotes
• make sure it starts with cloudinary://

This repo is already prepared for Render deployment.

Included deployment files:

• render.yaml
• build.sh
• Procfile

Render build flow:

pip install -r requirements.txt
python manage.py collectstatic --noinput
python manage.py migrate
python manage.py sync_media_to_cloudinary

Recommended production setup:

1. Connect the GitHub repository to Render.
2. Use Python 3.11.9.
3. Add CLOUDINARY_URL.
4. Add GROQ_API_KEY.
5. Add DATABASE_URL if you want persistent production data beyond SQLite.
6. Deploy from the main branch.

Useful local checks:

python manage.py check
python manage.py test account

Current automated coverage includes:

• revenue calculations
• Ask to AI endpoints
• session reset behavior
• production media fallback behavior

EcomCRM/
|-- account/
|   |-- management/
|   |-- templatetags/
|   |-- ai_assistant.py
|   |-- forms.py
|   |-- models.py
|   |-- tests.py
|   |-- urls.py
|   `-- views.py
|-- db/
|   |-- settings.py
|   |-- urls.py
|   `-- wsgi.py
|-- static/
|-- template/
|   |-- account/
|   |-- main.html
|   `-- navbar.html
|-- build.sh
|-- Procfile
|-- render.yaml
`-- requirements.txt

This repository demonstrates the ability to:

• ship an end-to-end Django product, not just isolated pages
• design a cleaner, more professional user experience without removing business features
• integrate third-party services like Groq and Cloudinary
• debug production deployment issues and harden the app for Render
• build features around real business metrics rather than placeholder data
• preserve operational workflows while upgrading architecture and presentation

EcomCRM is a strong portfolio project for full-stack, backend, product engineering, and AI-integrated web roles because it shows more than coding ability. It shows product judgment, system integration, deployability, UI execution, and resilience under real-world edge cases.