Shashi
Kathi
Data Analyst
Download Resume
Impact by Numbers
Real results from real projects
Projects Completed
Delivering actionable business insights
Data Models Built
Optimizing predictive performance
Accuracy Rate
In fraud detection & churn models
Hours of Analysis
Translating raw data into strategy
About
Me
I am a business-centric Data Analyst who bridges the gap between raw data and strategic decisions. My passion lies in solving complex problems—whether it's predicting customer churn or detecting fraud—to drive measurable efficiency and growth.
Beyond just technical skills in Python, SQL, and Power BI, I bring a focus on clarity, accuracy, and impact. I hold a strong foundation in cloud-based ML workflows and experience with AI-driven analytics.
My goal is not just to analyze data, but to translate it into actionable business insights that enhance forecasting, optimize operations, and empower stakeholders to make confident decisions.
Tools & Technologies
Hover to pause
Technical Skills
Languages & Libraries
Python (Pandas, NumPy, Scikit-learn), SQL, R
Data Visualization
Power BI, Tableau, Looker Studio, Matplotlib, Seaborn
Cloud & AI
Azure AI, GCP Vertex AI, Oracle Cloud, OpenAI APIs
Data Engineering
ETL Pipelines, Data Cleaning, Predictive Modeling
Experience
Data Science Research Intern
TransOrg Analytics | Remote
Jan 2025 – June 2025
- Cleaned and validated over 10,000 customer records.
- Built automated ETL pipelines using Python & SQL (30% faster reporting).
- Performed EDA to identify outliers and missing values.
- Developed reusable Python scripts for automation.
Education
Bachelor of Technology (Honors)
Lovely Professional University, Punjab
Aug 2021 – Aug 2025
Computer Science, Data Science & Data Engineering
Awarded 50% merit scholarship (AIR under 500)
Certifications
Oracle Cloud Infrastructure 2025 Data Science Professional
Google Cloud: Introduction to Data Analytics
Google: Looker Studio
Selected Work
Real-world projects with measurable impact
Financial Fraud Detection System
Problem: Financial fraud costs millions and requires immediate detection.
Solution: End-to-end ML pipeline with LightGBM and interactive Streamlit dashboard.
Outcome: Achieved 98.4% accuracy with real-time explainable insights.
Customer Churn Prediction
Problem: High customer turnover reduces revenue and profitability.
Solution: Predictive pipeline with XGBoost, SMOTE balancing, and live Hugging Face deployment.
Outcome: Identified at-risk customers with >85% accuracy for proactive retention.
Wine Quality Prediction
Problem: Subjective quality assessment lacks consistency and scale.
Solution: Random Forest model analyzing physicochemical properties.
Outcome: Identified key quality drivers with 77% accuracy for objective rating.
Let's Work
Together
I'm always open to discussing new projects, creative ideas, or opportunities to be part of your vision.