Hi, I'm Mary John.
A
I am ready-to-learn inquisitive individual, with a passion for applying advanced data analytics in real world. My strong academic background in Statistics, Economics and Mathematics as well as hands-on experience with machine learning and Artificial Intelligence projects especially in the finance and supply chain sectors empower me to solve complex problems and drive innovation.
About
I am a Business Analytics graduate student at Arizona State University with a focus on supply chain optimization and data-driven decision-making. I enjoy solving real-world problems using data and technology, and I always bring 100% energy, curiosity, and ownership to the work I do. I’ve worked with tools like Python, SQL, R, Tableau, Power BI, and Excel, and I hold multiple certifications on SAP BTP and SAP’s digital supply chain tools. My academic and industry experience has allowed me to apply machine learning, forecasting models, and process automation to real challenges—whether it’s optimizing logistics, improving customer returns, or analyzing operational data. I’m especially passionate about building intelligent systems that streamline operations, reduce inefficiencies, and improve user experience—from reverse logistics to AI-based forecasting.
- Languages: Python, R Programming, SQL, LaTeX
- Databases: MySQL
- Libraries: NumPy, Pandas, OpenCV
- Frameworks: Node.js, Keras, TensorFlow, PyTorch, Bootstrap
- Tools & Technologies: Power BI, Tableau, Excel, SPSS, Google Suite, Microsoft Office, AWS (Amazon EC2, Amazon Bedrock)
- SAP & Certifications: SAP BTP (Certified), SAP IBP, Data Analysis with Python (IBM), Power BI (Microsoft)
Looking for an opportunity to work in a challenging position combining my skills in Analytics, Statistics, Economics and Supply Chain, which provides professional development, interesting experiences and personal growth.
Experience
- Worked on SAP Business Technology Platforms.
- Processed data and KPI filteration of various companies through Bloomberg and analysed through data visualizations using SAP Cloud Analytics.
- Improved the response time by 20% by refactoring and connecting Joule to S4/HANA as well as SAP Cloud Analytics for smoother client experience
- Performed various visulaizations on KPI's to help in creation of the product called CFO Dashboard
- Tools: SAP Cloud Analytics, SAP BTP, SAP S4/HANA, Bloomberg, Microsoft Excel, Google Suite, SQL
- Engineered a robust Explainable AI framework for early-stage breast cancer detection in collaboration with HonorHealth, enabling transparent clinical decision-making.
- Utilized CVAT for precise annotation of breast tissue anomalies, including mass formations and architectural distortions, to support classification into benign and malignant categories.
- Deployed YOLOv8-based object detection integrated with Ultralytics for high-resolution image segmentation; achieved 96% classification accuracy by optimizing hyperparameters, fine-tuning backbone architectures, and performing cross-validation on expert-annotated datasets.
- Coordinated with dedicated annotation, modeling, and validation teams to ensure streamlined workflows across dataset preparation, model training, and clinical evaluation pipelines.
- Tools: Python, CVAT, YOLOv8, Ultralytics, PyTorch, OpenCV, TensorBoard, NumPy, Pandas, Scikit-learn.
- Verified and analyzed financial data for over 1,000 companies using Excel tools (Pivots, Macros) to identify risk patterns and optimize pricing models, leading to a 20% reduction in premium discrepancies.
- Enhanced renewal strategies and actuarial pricing models by analyzing claims data and market trends, contributing to a 10% increase in pricing accuracy and surpassing ₹300 crore in premium collection.
- Streamlined workflows for death benefit covers, including Terminal Illness, Critical Illness, Accidental Death Benefit, and Permanent Total Disability add-ons, ensuring compliance with Turn-Around Time (TAT) protocols and improving operational efficiency by 15% using QLik Sense to monitor progress.
- Tools: Advanced Microsoft Excel, QLik Sense, Microsoft Word, Microsoft Outlook
Projects
A facial recognition solution for streamlining ASU graduation ceremonies using real-time CV.
- Tools: Python, OpenCV, dlib, face_recognition, Amazon EC2, venv
- Developed a real-time facial recognition system to replace paper-based QR codes at ASU graduation ceremonies.
- Automated student check-in, photo tagging, and Jumbotron display through facial identification.
- Deployed and tested the solution on Amazon EC2 with over 97% accuracy under controlled conditions.
ML project analyzing Yelp reviews of Tucson pizza restaurants using NLP and recommendation systems.
- Tools: Python, Pandas, Matplotlib, BERT, Amazon Bedrock, LoRA
- Performed sentiment analysis using BERT to extract opinions on aspects like crust, cheese, service, and ambiance..
- Used Claude (Amazon Bedrock) for auto-generated review summaries.
- Built a LoRA-powered recommendation model to suggest top 3 restaurants based on user-specified keywords.
- Delivered actionable insights for both customers and business owners.
An optimization project maximizing profits for a small custom calendar business using linear programming.
- Tools: Linear Programming, MS Excel Solver, Scenario Analysis, Product Mix Strategy
- Developed models to maximize profit under production time, capacity, and demand constraints.
- Analyzed multiple scenarios; recommended diversified product mix with Basic, Portrait, and Resin calendars.
- Achieved profit of $631 while meeting business goals and resource limits.
Lean Six Sigma case study to reduce proposal cycle time for a $60B global tech company.
- Tools: Tableau, MS Excel, Lean Six Sigma (DMAIC), ERP Data Analysis, ANOVA, Correlation Analysis
- Analyzed proposal creation delays across brands and regions using ERP timestamp data.
- Identified root causes: hand-offs, manual approvals, incomplete bid requests
- Recommended automation, SOP enforcement, and standardized templates.
- Proposed solutions to cut proposal cycle time by 15% and improve process efficiency.
Machine learning analysis of factors contributing to depression using real-world data.
- Tools: Python, MS Excel, Jupyter Notebook, Pandas, NumPy, Scikit-Learn, Seaborn, Matplotlib
- Cleaned and analyzed data to identify patterns in lifestyle, health, and family dynamics.
- Built and evaluated models (Logistic Regression, Decision Trees) to predict depression risk.
- Achieved strong model accuracy and extracted actionable insights for mental health awareness.
Analyzed Fitbit data to predict calorie burn and extract fitness insights using machine learning.
- Tools: Python, Jupyter Notebook, Pandas, NumPy, Scikit-Learn, Matplotlib, Seaborn
- Performed EDA on activity, sleep, and calorie data to identify trends and behaviors.
- Built multiple classification models (Logistic Regression, KNN, Naive Bayes, Decision Tree, Random Forest).
- Random Forest achieved 78.98% accuracy in predicting calorie burn levels.
- Provided actionable recommendations to optimize fitness routines and lifestyle habits.
Analyzed World Happiness Report data to evaluate key factors impacting happiness across Liberia, Ukraine, and Congo.
- Tools: Python, Jupyter Notebook, Pandas, NumPy, Scikit-Learn, Statsmodels, Matplotlib, Seaborn
- Conducted correlation analysis to study relationships between happiness score and GDP, social support, life expectancy, freedom, generosity, and corruption.
- Built a multiple linear regression model to quantify the influence of each factor on the happiness score.
- Provided insights into how external factors like geopolitical conflict may impact happiness beyond economic indicators.
Skills
Languages and Databases
Python
HTML5
CSS3
MySQL
PostgreSQL
Shell Scripting
Libraries
NumPy
Pandas
OpenCV
scikit-learn
matplotlib
Frameworks
Django
Flask
Bootstrap
Keras
TensorFlow
PyTorch
Other
Git
AWS
Heroku
Education
Tempe, USA
Degree: Master of Science in Business Analytics
CGPA: 3.93/4.0
- AI in Business
- Descriptive & Predictive Analytics
- Business Process Analytics
- Analytical Decision Modelling
- Enterprise Data Analytics
- Machine Learning in Business
- Advanced Marketing Analytics
- Strategic Procurement
Relevant Courseworks:
Bengaluru, India
Degree: Master of Science in Statistics
CGPA: 3.23/4.0
- Design of Experiments
- Statistical Quality Control
- Time Series Analysis
- Neural Network and Deep Learning
- Probability Theory
- Statistical Inference
- Principles of Data Science and Database Techniques
- Regression Analysis
Relevant Courseworks:
Bengaluru, India
Degree: Bachelor of Science Economics, Mathematics, Statistics (Triple Major)
CGPA: 3.69/4.0
- Macroeconomics and Microecomics
- Algebra
- Fiancial Economics, Mathematics
- Complex Analysis, Real Analysis
- Enterprise Data Analytics
- Differential Equations
- Sample Survey Designs
- Econometrics and International Economics
Relevant Courseworks: