A full-stack analytics solution seamlessly deployed and running adjacent to the data inside the Snowflake Data Cloud. This project integrates real-time operational data with Generative AI to provide executives with immediate, actionable insights into platform profitability, restaurant performance, and customer loyalty.
This project directly addresses the need for real-time, granular strategic decisions in the high-velocity food delivery industry. It showcases an innovative, enterprise-grade architecture where analytics are performed where the data lives, eliminating latency and data transfer costs.
The primary goal is to unlock actionable intelligence to inform strategic decisions:
- Profitability Optimization: Identify top-grossing restaurants, commission revenue, and net profit to optimize commission rates and menu offerings.
- Customer Loyalty: Track customer demographics, join dates, and order frequency to develop targeted marketing campaigns and loyalty programs.
- Operational Excellence: Analyze ratings, cuisine types, and commission rates to inform restaurant selection and menu optimization.
By leveraging these insights, users can: Optimize menu offerings and pricing strategies, enhance customer retention and loyalty programs, and inform strategic business decisions to drive growth and profitability.
This project leverages a cutting-edge, integrated suite of tools for secure and efficient analytics.
| Component | Technology | Showcase Skill / Purpose |
|---|---|---|
| Application Hosting | Streamlit in Snowflake (SiS) | Demonstrates zero data egress, high performance, and automatic scaling within the data cloud perimeter. |
| Data Engine | Snowflake Data Cloud | Unified platform for data warehousing, compute, and application execution. |
| NL2SQL / AI | Snowflake Cortex AI | Implements secure Natural Language Query functionality (Tab 4) and dynamic text summarization (Tab 1). |
| Frontend/Logic | Python 3.9+ | Used for dashboard logic, data manipulation, and utilizing Streamlit/Plotly libraries. |
The portal provides a multi-dimensional view designed for strategic navigation:
| Tab | Icon | Core Focus |
|---|---|---|
| 1. Portal Summary | 📰 | AI-Generated Context explaining the dashboard's objective and core data structure. |
| 2. Executive Dashboard | 🚀 | High-level financial (GMV, Profit) and operational KPIs. |
| 3. Restaurant Deep Dive | 🍽️ | Analysis of profitability, commission impact, and rating correlation. |
| 4. AI Analyst | 🧠 | NL2SQL Engine: Ask business questions; Cortex generates and executes SQL live. |
| 5. Customer Insights | 👥 | Metrics on customer loyalty, repeat rate, and satisfaction distribution. |
| 6. Conclusion & Recs | 🧩 | Executive Action Plan combining insights into data-backed next steps. |
This section highlights the specific software engineering and data governance skills demonstrated in the project:
-
NL2SQL Integration: Designed and implemented a secure AI workflow (Tab 4) where user text is dynamically passed to
SNOWFLAKE.CORTEX.COMPLETE()to instantly generate and execute complex SQL queries against the underlying tables. -
Strategic Conclusion Synthesis: Developed logic (Tab 6) to run multiple aggregate queries and use Python conditional statements (
if/then) to generate contextual, data-driven strategic recommendations. -
Data Model & Source: Built upon a clear structure encompassing six core files (e.g.,
FACT_ORDER_STREAMLIT.csv,DIM_RESTAURANT_STREAMLIT.csv) for a comprehensive view of the business. -
Custom UI & State Management: Utilized Streamlit's
st.session_statefor managing custom features like the Dark/Light Mode toggle and complex filter states. -
Secure Git Synchronization: Established a secure connection between the internal Snowflake environment and this public GitHub repository, demonstrating proficiency in cloud-native version control and CI/CD preparation via Snowflake's Git Integration DDL.
The repository is structured to enable immediate deployment within the Streamlit in Snowflake environment. It contains the following core files, with the CSV files simulating the data tables:
| Source File | Type | Purpose |
|---|---|---|
FACT_ORDER_STREAMLIT.csv |
FACT | Core order transactions, GMV, and profit metrics. |
DIM_CUSTOMERS_STREAMLIT.csv |
DIM | Customer demographics and loyalty metrics. |
DIM_RESTAURANT_STREAMLIT.csv |
DIM | Restaurant details, ratings, cuisine type, and commission rates. |
DIM_FACT_ORDER_ITEMS_STREAMLIT.csv |
FACT | Detailed breakdown of items within each order. |
DIM_MENU_ITEM_STREAMLIT.csv |
DIM | Menu category and item pricing details. |
DIM_COUPON_STREAMLIT.csv |
DIM | Coupon usage and discount information. |
streamlit_app.py: The core application logic.environment.yml: Defines Python dependencies.Data/(Folder): Contains the six core CSV data files.
Example: 