Pragathi Porawakara Arachchige

Hi, I'm Pragathi

Lead Data Scientist | AI/ML Expert | Data Strategy Leader

Dynamic Lead Data Scientist with 8+ years of experience in AI-driven solutions, machine learning, and data governance. Proven track record of improving operational efficiency by 40%, reducing manual intervention by 50%, and enhancing decision-making through predictive analytics.

Lead team of junior data scientists
Pragathi Profile Photo

Professional Impact

40%

Operational efficiency improvement

50%

Reduction in manual processes

20%

Increase in customer engagement

$1.5M

Annual revenue retained

Featured Projects

Telecom Customer Churn Prediction

Built a machine learning model using ensemble techniques to predict telecom customer churn with 92% ROC-AUC, enabling proactive retention strategies.

Python SBERT NLP Scikit-Learn

Call Center Performance Analysis

Developed predictive models using Ridge Regression and Random Forest to analyze call center data, achieving 98% accuracy in identifying high abandonment scenarios and guiding operational improvements.

Ridge Regression Random Forest Classifier Residual Analysis Correlation

Topic Modeling for BBC News

Implemented LDA topic modeling to categorize custmer news, improving efficiency by 15% and reducing handling time by 10%.

LDA NLP Python Scikit-Learn

Sentiment Analysis BBC News

NLP pipeline for sentiment classification using VADER and TextBlob with comparative analysis.

Python NLP TextBlob VADER

FIFA21 Age vs Performance Analysis

Exploratory Data Analysis of FIFA21 player dataset examining the relationship between age and performance metrics with interactive visualizations.

Python Pandas Seaborn Matplotlib EDA

Call Center Sentiment Analysis

NLP analysis of customer service call transcripts to identify sentiment patterns and improve service quality using VADER and TextBlob.

Python NLP TextBlob VADER Pandas Matplotlib

Soccer Team Performance Analysis

Comprehensive analysis of soccer team metrics with interactive dashboards to evaluate team strategies and player contributions.

Python Pandas Tableau Seaborn Sports Analytics
Combines professional skills with personal passion

Credit Card Fraud Detection

Machine learning system to detect fraudulent transactions with 99% accuracy using anomaly detection algorithms.

Python Scikit-Learn Isolation Forest SMOTE Feature Engineering

Call Volume Forecasting

ARIMA and Prophet models to predict call center volumes, reducing staffing costs by 15% through optimized scheduling.

Python ARIMA Prophet Time Series Statsmodels

Player Value Prediction

Predicting soccer player transfer values using performance metrics with 85% accuracy (RMSE: €1.2M).

Python XGBoost Feature Engineering Sports Analytics Web Scraping
Real Madrid transfer target analysis included
View All Projects on GitHub

Technical Skills

Machine Learning

Scikit-Learn TensorFlow XGBoost Random Forest SBERT TF-IDF LDA Jupyter

Data Engineering

Informatica Azure Data Factory AWS S3 ETL Pipelines Kafka Snowflake Redshift

Databases

SQL Server Oracle PostgreSQL MongoDB

Data Visualization

Tableau Power BI Spotfire Tableau Prep SSRS

Tools & Platforms

GitHub VS Code Alteryx Splunk Excel (Advanced)

Operating Systems

Windows Linux

Programming

Python SQL Bash JavaScript

Cloud Platforms

AWS Azure GCP SageMaker

MLOps

Docker CI/CD Kubeflow MLflow

Professional Experience

2019 - Present
Lead Data Scientist / Lead Data Engineer
Charter Spectrum, Saint Louis, MO
  • Led team of 5 data scientists in developing AI solutions that improved customer service efficiency by 40%
  • Built vector search-based recommendation engine increasing customer engagement by 20%
  • Optimized ETL pipelines improving data ingestion efficiency by 40%
2018 - 2019
Senior IT Data Modeling Lead Analyst
Express Scripts, Saint Louis, MO
  • Improved query performance by 35% through SQL Server data modeling
  • Reduced data processing time by 50% through ETL pipeline optimization
  • Increased data retrieval efficiency by 40% through architecture redesign
2017 - 2018
Business Data Operations Analyst IV
Wells Fargo Advisors, Saint Louis, MO
  • Automated reporting workflows reducing manual effort by 40%
  • Improved forecasting accuracy by 20% through advanced statistical analysis
  • Enhanced data quality by 30% through governance standards

About Me

Dynamic Lead Data Scientist with 8+ years of experience in AI-driven solutions, machine learning, and data governance. Proven track record of improving operational efficiency by 40%, reducing manual intervention by 50%, and enhancing decision-making through predictive analytics and data visualization.

I hold three Master's degrees (two MSc and one MBA) which have equipped me with both technical expertise and business acumen. My technical specialties include:

Machine Learning Natural Language Processing Data Engineering Cloud Architecture Data Visualization

Outside of work, I'm an avid Real Madrid fan, world traveler, and enthusiast of Asian dramas. These diverse interests help me maintain creativity and bring unique perspectives to problem-solving in data science.

Get in Touch

I'm currently open to discussing new opportunities, collaborations, or speaking engagements. Feel free to reach out via any of these channels: