T his project started as an idea to process 1.3 million rows of financial data interactively. Over time, it evolved into a powerful AI-powered web application that can accept any size dataset, clean it, perform feature engineering, and visualize it in an intuitive UI—all powered by GPT-4o, LangChain, and serverless AWS tools.
🌐 Live Architecture
- EC2: Flask backend with Gunicorn and Nginx
- S3: Secure file storage for datasets
- RDS (MySQL): Stores metadata and session logs
- Lambda: Triggers OpenAI agents for data cleaning
- Secrets Manager: Secure API key management
- GitHub Actions: Handles CI/CD deployment
Backend
Flask
Python
Gunicorn
Frontend
HTML
Streamlit
Jinja2
TailwindCSS
AI Agents
GPT-4o
AI Data Science Team
LangChain
Infrastructure
S3
EC2
RDS
Lambda
Secrets Manager
CI / CD
GitHub Actions
Let’s Connect
If you’re a builder, dreamer, or data explorer, reach out. Whether it's a conversation, a collaboration, or just curiosity, I’d love to connect. You can find me on LinkedIn or email me directly.
🧪 How It Works
Step 1: Register / Login
Access the Dashboard
New users create an account or log in via a secure Flask authentication flow. Once authenticated, they land in the main dashboard.
Step 2: Upload CSV
Flask Handles Upload to S3
Users drag and drop datasets directly into the dashboard. Files are uploaded to AWS S3 using secure presigned URLs to bypass local limitations.
Step 3: Clean + Engineer
Lambda + AI Agents
Users trigger an AI-powered pipeline that runs `DataCleaningAgent` and `FeatureEngineeringAgent` on the dataset asynchronously using AWS Lambda.
Step 4: Store Output
Cleaned Data in S3 + Logs in RDS
Once processed, cleaned files are saved in a `cleaned/` prefix in S3. All job metadata, timestamps, and session info are stored in RDS for traceability.
Step 5: Explore with AI
Streamlit-Powered EDA
Users launch a dynamic Streamlit interface that loads the cleaned preview and visual insights. Features include charts, agentic mind maps, and NL → SQL querying.
"This is super cool! I remember struggling to bridge that gap between insights and code. A tool that shows the 'how' is a total game-changer for learning." - Tran Tien Van
📁 Repository Structure
Cloudberry_AWS_Bootcamp/
│
├── Portfolio_V2/ # Core Flask app
│ ├── app.py # Entrypoint Flask application
│ ├── templates/ # Jinja2 HTML templates
│ ├── static/ # TailwindCSS and JS
│ ├── utils/ # AI pipeline, S3 handlers, and secrets manager
│ ├── data_forge_lite/ # Streamlit-powered exploration dashboard
│ └── requirements.txt
│
├── .github/workflows/ # GitHub Actions CI/CD scripts
├── README.txt # You're reading it!
└── start.sh # Launch script for Gunicorn
📁 Run Locally
git clone https://github.com/LannonTheCannon/Cloudberry_AWS_Bootcamp.git
cd Cloudberry_AWS_Bootcamp/Portfolio_V2
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
python3 app.py