📊 Data Analyst Portfolio

Peddireddy Soma
Siva Saran Teja

Transforming raw data into actionable insights through Python, SQL, and Power BI — across banking, e-commerce, retail, and energy domains.

PythonSQL · MySQLPower BI EDAMachine LearningData VisualisationExcel
View Projects → GitHub Profile Get in Touch
5Projects
6GitHub Repos
6Certifications
7.99B.Tech CGPA

🐙 All source code available on GitHub — 6 public repositories

github.com/sarantej1
Where I've Worked

Experience

Hands-on industry internship delivering end-to-end data analytics projects

Data Science Intern
SmartED Innovations — USNG Learnspace EdTech Pvt. Ltd., Bangalore
Oct 2024 – Mar 2025
Certificate ID: DS1147
✦ Verified Internship · DS1147
  • Developed a Customer Shopping Behavior Analysis Power BI dashboard on 99,457 mall transactions, segmenting by gender, age group, category & payment method.
  • Built an end-to-end Blinkit Grocery Sales Analysis in Power BI, tracking ₹1.2 Cr+ in sales KPIs, outlet performance, and item-type profitability across tier cities.
  • Applied Python (Pandas, Seaborn, Sklearn) for EDA and KMeans clustering on retail datasets with 100K+ records.
  • Automated cleaning pipelines — missing values, outlier detection, feature engineering — cutting manual effort ~60%.
⭐ Internship Projects Delivered
📊 Internship Project 1
Customer Shopping Behavior Intelligence Dashboard
Power BI dashboard on 99,457 transactions across 10 Istanbul malls — deep patterns in age, gender, category preference & payment behaviour.
Power BIPythonPandasSeaborn
🔗 github.com/sarantej1/Customer-Shopping-Behavior-Intelligence-Dashboard
🛒 Internship Project 2
Blinkit Grocery Sales Performance Analytics
Executive Power BI dashboard tracking ₹1.2 Cr+ in Blinkit grocery sales — outlet types, item-category profitability, city-tier KPIs & DAX measures.
Power BIDAXData Modelling
🔗 github.com/sarantej1/Blinkit-Grocery-Sales-Performance-Analytics
What I've Built

Projects

Click "View Files" on any card to explore datasets, code & reports — or open GitHub directly

Finance & Banking · SQL
Banking Transaction Intelligence System

End-to-end SQL analysis of 100 Indian banking transactions across 40 customers and 15+ cities — spending patterns, risk indicators, and customer lifetime value.

📊₹8.2 Cr+ transaction volume analysed (Jan–Apr 2024)
🔍25+ SQL queries: CTEs, Window Functions (LAG, RANK, SUM OVER)
⚠️Anomaly detection: 1 failed txn + 3 high-value outliers flagged
👥Age segmentation: 31–45 mid-career = 48% of total volume
MySQL 8.0Window FunctionsCTEsSubqueries
3 project files
GitHub
E-Commerce · Python + Power BI
E-Commerce Revenue & Customer Analytics Platform

Full EDA and ML on 10,000 orders (31 features, 2022–2024) across 8 product categories, culminating in a Power BI executive dashboard.

💰Electronics top revenue: ₹9.09 Cr; Smartphones highest sub-category
🧹Imputed 2,457 missing records; 6 engineered features (AOV, profit_segment)
📈Heatmap: unit_price→tax (0.797), revenue→COGS (0.984)
PythonPandasSeabornSklearnPower BI
3 project files
GitHub
Energy & Sustainability · Time-Series ML
EV Charging Demand Forecasting Using ARIMA Models

Time-series forecasting predicting future EV charging demand to enable data-driven placement and capacity planning of charging infrastructure.

⚡ARIMA (p,d,q) models on historical EV charging demand time series
📉ACF/PACF analysis for optimal hyperparameter selection
🗺️Forecasts guide optimal station placement & grid capacity planning
PythonStatsmodels ARIMAPandasMatplotlib
Academic Capstone · SRM
GitHub
Retail Analytics · Power BI · ⭐ Internship
Customer Shopping Behavior Intelligence Dashboard

Power BI dashboard analysing 99,457 shopping transactions across 10 Istanbul malls — revealing patterns in age, gender, category preference and payment behaviour.

🛍️99,457 transactions · 10 malls · 8 categories · 3 payment methods
👩Female ~60K vs Male ~40K; Clothing dominated all categories
💳Cash led total sales; Mall of Istanbul & Kanyon top malls
Power BIPythonPandasSeaborn
SmartED Internship · 2 files
GitHub
Grocery & Q-Commerce · Power BI · ⭐ Internship
Blinkit Grocery Sales Performance Analytics

Executive-grade Power BI dashboard tracking ₹1.2 Cr+ in Blinkit grocery sales — item categories, outlet types, city tiers, and profitability KPIs.

🛒₹1.2 Cr+ total sales · 8,523 items · Avg Rating 3.9/5
📦Top items: Fruits & Veg, Snack Foods, Household
🏙️Tier-3 cities lead; Medium outlets = highest revenue
Power BIDAXData Modelling
SmartED Internship · 1 file
GitHub
Technical Arsenal

Skills

Tools used across real-world data projects

🐍 Programming
PythonPandasNumPyMatplotlibSeabornScipy
🗄️ Databases & SQL
MySQL 8.0DDL / DMLWindow FunctionsCTEsSubqueries
📊 BI & Visualisation
Power BIDAXExcelDashboards
🤖 Machine Learning
SklearnKMeansLinear RegressionARIMAFeature Engineering
🧹 Analytics
EDAData CleaningCorrelation AnalysisOutlier DetectionTime-Series
💼 Soft Skills
Attention to DetailProject ManagementDecision MakingTeamwork
Academic Background

Education

Strong foundation in computer science and engineering

B.Tech — Computer Science & Engineering
SRM Institute of Science and Technology, Chennai
Sep 2020 – May 2024CGPA: 7.99 / 10
Class XII — MPC
Narayana Junior College, Vijayawada
Jun 2018 – Apr 2020CGPA: 8.84 / 10
Class X
Narayana School, Kakinada
Jun 2017 – Apr 2018CGPA: 9.8 / 10
Credentials

Certifications

Industry-recognised credentials from leading platforms

🌐
Introduction to Data Science
Cisco Networking Academy
📊
Microsoft Power BI
Infosys Springboard
📗
Data Analyst 101: Excel Formulas & Functions
Simplilearn
🐍
Python Programming
Skill India
🗄️
Database Foundations
Oracle Academy
💻
SQL
HackerRank
Get In Touch

Let's Connect

Open to data analyst roles, freelance projects, and collaborations.

✉️
Email
sarantej333@gmail.com
📞
Phone
+91 9392428829
📍
Location
Kakinada, Andhra Pradesh
GitHub

sarantej1

Peddireddy Soma Siva Saran Teja

6 Public Repos5 Data Projects
Visit GitHub
🏦 Banking Transaction Intelligence System

📄 banking_analysis.sql

Complete MySQL script — 8 sections, 25+ queries covering transaction trends, customer segmentation, merchant analysis, city performance, window functions, CTEs, and risk detection.

MySQL 8.0~200 lines8 Sections
View on GitHub
-- Banking Transaction Intelligence System | MySQL 8.0 CREATE TABLE bank_transactions ( transaction_id VARCHAR(10) PRIMARY KEY, customer_id VARCHAR(10), age INT, city VARCHAR(50), amount DECIMAL(15,2), status VARCHAR(20) ); -- Month-over-Month Growth (Window Function) WITH monthly_volume AS ( SELECT DATE_FORMAT(transaction_date,'%Y-%m') month, ROUND(SUM(amount),2) total_volume FROM bank_transactions WHERE status='Success' GROUP BY month ) SELECT month, total_volume, LAG(total_volume) OVER(ORDER BY month) prev_month, ROUND((total_volume - LAG(total_volume)OVER(ORDER BY month)) * 100 / NULLIF(LAG(total_volume)OVER(ORDER BY month),0), 2) AS mom_growth_pct FROM monthly_volume; -- Jan:₹18.5L Feb:₹19.8L Mar:₹30.5L Apr:₹24.2L -- Customer Lifetime Value (CTE) WITH customer_stats AS ( SELECT customer_id, customer_name, COUNT(*) num_transactions, ROUND(SUM(amount),2) total_volume, DATEDIFF(MAX(transaction_date),MIN(transaction_date)) lifespan_days FROM bank_transactions WHERE status='Success' GROUP BY customer_id, customer_name ) SELECT *, ROUND(total_volume/NULLIF(lifespan_days,0)*365,2) annualised_value FROM customer_stats ORDER BY total_volume DESC;
🛒 E-Commerce Analytics Platform

📓 Untitled16.ipynb — Full EDA Notebook

Python EDA: null handling (2,457 records imputed), 6 engineered features (profit_segment via qcut, delivery_performance, AOV), correlation heatmap, category revenue & sub-category profit ranking.

Jupyter Notebook10,000 rows31 features
View on GitHub
import pandas as pd; import seaborn as sns df = pd.read_excel("ecommerce_dataset.xlsx") # (10000, 31) # Null imputation (2,457 missing records fixed) df["delivery_days"].fillna(df["delivery_days"].mean(), inplace=True) df["customer_rating"].fillna(df["customer_rating"].mean(), inplace=True) # Feature Engineering df["profit_segment"] = pd.qcut(df["profit"], q=4, labels=["Low","Medium","High","Very High"]) df["delivery_performance"] = pd.cut(df["delivery_days"], bins=[0,2,5,10], labels=["Fast","Normal","Slow"]) df["average_order_value"] = df["total_revenue"] / df["quantity"] # Revenue by Category df.groupby("category")["total_revenue"].sum() # Electronics: ₹9,09,43,230 ← Highest # Home&Kitchen: ₹2,56,40,025 # Correlation Heatmap (key findings) # unit_price → tax_amount : 0.797 # total_revenue → cogs : 0.984
⚡ EV Charging Demand Forecasting

⚡ EV-Charging-Demand-Forecasting-Using-ARIMA-Models

Academic capstone from SRM Institute of Science & Technology. Forecasts EV charging demand using ARIMA time-series modelling on historical infrastructure data.

Jupyter NotebookTime-Series MLAcademic Capstone
View on GitHub

Project Objectives

• Forecast EV charging demand using ARIMA (p,d,q)
• Optimise hyperparameters via ACF/PACF analysis
• Validate on held-out test sets for reliability
• Inform station placement and grid capacity planning
• Enable sustainable transportation network development

🛍️ Customer Shopping Behavior Dashboard

📓 Untitled15.ipynb — EDA Notebook

EDA on 99,457 mall transactions: gender/age distribution, category analysis, payment method breakdown, shopping mall rankings, and total sales by year. Zero nulls in raw data.

Jupyter Notebook99,457 rowsSmartED Internship
View on GitHub
import pandas as pd; import seaborn as sns df = pd.read_csv("customer_shopping_data.csv") # Shape: (99457, 10) — Zero nulls! df['total_sales'] = df['quantity'] * df['price'] df['age_group'] = pd.cut(df['age'], bins=[18,25,35,45,60,80], labels=["18-25","26-35","36-45","46-60","60+"]) # Age groups : 46-60 leads (28,434) | 36-45 (19,436) # Gender : Female ~60K | Male ~40K transactions # Top Category: Clothing (35,000+ purchases) # Top Mall : Mall of Istanbul > Kanyon > Metrocity # Payment : Cash leads total sales volume
🛒 Blinkit Grocery Sales Analytics

📊 BLINKIT.pbix — Power BI Dashboard

Executive Power BI report: total sales, avg ratings, item visibility, outlet performance across fat-content types, city tiers, and outlet sizes. Delivered during SmartED internship.

Power BI (.pbix)SmartED Internship₹1.2 Cr+ Sales
Download from GitHub

Dashboard Visuals

• KPI Cards: Total Sales, Avg Sales/Item, No. of Items, Avg Rating
• Fat Content donut: Regular vs Low-Fat split
• Item Type bar: Fruits & Veg, Snack Foods, Household top 3
• Outlet Size: Medium = highest revenue
• Location: Tier-3 cities lead sales
• Outlet Type matrix: Supermarket Type 1 dominates

8,523 ItemsAvg Rating 3.9DAX Measures