About
With a Master’s degree in Business Analytics and nearly four years of experience in financial services and healthcare consulting, I excel at data-driven decision-making and business innovation.
- At American Express, I led risk mitigation strategies for digital banking and payment products, significantly reducing credit and fraud risks.
- At ZS Associates, I turned complex data into actionable insights, enhancing process automation and efficiency.
- Passionate about employing technical expertise and strategic leadership to create meaningful change and deliver outstanding outcomes with innovative technological solutions.
- Comprehensive skill set in Product Management, Data Analytics, and Data Science.
- Website: www.vineethpydi.com
- LinkedIn: linkedin.com/in/vineethpydi
- Email: [email protected]
- Phone:
- City: Chicago, IL / West Lafayette, IN
- Degree: MS in Business Analytics and Information Management
SKILLS
Tools & Techniques
- Python (PyTorch, PySpark, Scikit-learn, TensorFlow)
- SQL
- Tableau
- SAS
- BigQuery, Vertex AI
- R
- Agile - Scrum, Kanban
- Product Life-cycle Management
Software & Plaftforms
- Cloud - Google Cloud, AWS, MS Azure
- Jira, Smartsheets, MS Projects
- Confluence
Modelling/Statistics
- Data Mining, Text Mining
- Machine Learning
- ANOVA, p-test, f-test
- A/B Testing, Hypothesis Testing
- Regression, Clustering, Classification
- Time Series Forecasting
Business/Leadership
- Agile - Scrum, Kanban
- Product/Project Management
- Stakeholder Management
EDUCATION
Purdue University Daniels School of Business West Lafayette, Indiana, USA
Master of Science in Business Analytics and Information Management August 2023 - July 2024
Relevant Coursework: Data Mining, Big Data (GCP), Cloud Computing (AWS), AI in Business, Data Science in the Cloud (AWS), Visual Analytics, Optimization Modelling, Pricing Strategy, Competitive Strategy, Advanced Business Analytics
SRM Institute of Science and Technology Chennai, India
B.Tech. in Electronics and Communication Engineering March 2016 - March 2020
Relevant Coursework: Data Analytics using Python, C++, Python Programming, Data Visualization, Probability, Statistics, Project Management
Resume
View my ResumeSumary
Vineeth Pydi
Data-driven decision maker with about four years of experience empowering teams to build advanced models and products through analytical strategies and innovation. Skilled at forming cross-functional partnerships to translate business goals into strategies that drive business transformation and increase efficiency and revenue.
- West Lafayette, Indiana, USA
- +1 765 543 8774
- [email protected]
Education
Master of Science in Business Analytics & Information Management
Purdue University, Daniels School of Business
Aug 2023 - Jul 2024
• Graduate Teaching Assistant – Business Analytics, Federal Budgets
• Graduate Research Assistant – Department of Economics
• Projects: Time-Series Transformer Modelling (AI/ML), Data For Good (Azure AI Case), SAS Optimization Challenge (SAS), Bankruptcy Prediction (Python), Credit Risk Assessment (R), Stock Price Prediction (AWS)Bachelor of Technology in Electronics & Communication Engineering
SRM Institute of Science and Technology
Mar 2016 - Mar 2020
• Publications: Translation of Gesture-Based Static Sign Language to Text and Speech (IOPScience)
Certifications
Technical
- Certified Analytics Professional
- Data Scientist with Python
- AWS Certified Cloud Practitioner
- IBM Data Analyst Professional Certificate
- Tableau Desktop Specialist
- Operations research with SAS
- Snowflake LLM Bootcamp
Business/Leadership
- Harvard Leadership Edge - Storytelling with Data
Professional Experience
Data Science Consultant – Part-Time/Internship
Accenture Strategy, Chicago, IL
Jan 2024 - Present
- Spearheaded demand forecasting and feasibility analysis in the FMCG industry, utilizing transformer models for cutting-edge time series predictions and market sentiment analysis, thereby elevating strategic decision-making
- Pioneered the development and training of Transformer-based models, incorporating statistical analysis and AI/ML techniques to boost forecast accuracy by 25%, facilitating enhanced demand planning
Sr. Associate, Digital product Management
American Express
Jan 2022 - Jul 2023
- Led a 5-person Agile team in refining risk strategies for US digital banking and payment products by integrating advanced credit models, reducing risk exposure by 15%, and achieving annual savings of $1.2M
- Consolidated and integrated external credit bureaus and APIs to enhance risk controls, boosting model accuracy by 18%, resulting in the team receiving the Edward P. Gilligan Award for Innovation in 2023
- Enhanced risk strategies to reduce decision turnaround time by 25%, and played a pivotal role in the development and integration of Amex-acquired Kabbage portfolio, yielding a $600K pre-tax income benefit in 2023
- Collaborated with key stakeholders to implement advanced risk models, directly contributing to over $700K in annual savings on credit losses through improved decision-making processes
- Led Fastrack and PI initiatives from requirements to delivery, crafting user stories and UAT criteria, and streamlined the Agile process for improved efficiency and product delivery
Decision Analyst
ZS Associates
Jan 2020 - Jan 2022
- Collaboratively designed and executed an incentive compensation plan for a Fortune 500 US pharmaceutical client for quarterly sales of $2B+ among a 3-tiered 800+ sales force
- Spearheaded the migration of SAS to AWS, enhancing data and machine learning workflows by integrating AWS S3, Glue, Redshift, and SageMaker, achieving a 20% reduction in operational costs
- Analyzed COVID-19's impact, forecasting a 15% sales variance, leading to incentive plan adjustments that resulted in a 10% sales growth within the next quarter, thereby ensuring fairness and motivation
- Streamlined and automated key workflows, leading to a 20% reduction in manual effort and a 90% reduction in errors ensuring 100% accurate quarterly payouts to over 800 sales reps
PROJECTS
Application of Transformer Models for Demand Forecasting in FMCG Industry Code
Tech Stack: PyTorch, Deep Learning, Transformer-based models, Sentiment Analysis, Web Scraping, Data Mining
In the fast-paced FMCG industry, accurate demand forecasting is essential for competitive advantage and operational efficiency. Our research at Purdue University uses advanced transformer models, specifically iTransformer and TimesNet, to revolutionize demand forecasting by integrating traditional data with external sources like Google Trends and Amazon reviews. This approach enhances precision and reliability, addressing dynamic consumer preferences and providing actionable insights for inventory management and strategic planning. Supported by distinguished professors and an industry partner, this project demonstrates the power of combining state-of-the-art machine learning with comprehensive data analysis, significantly improving traditional forecasting methods. Due to a Non-Disclosure Agreement (NDA), specific data sets and the company name cannot be disclosed.
Enhanced Credit Risk (Loan) Decision System Code
Tech Stack: R, Machine Learning for classification and regression, XGBoost, Adaboost, Blackboost
This project employs advanced machine learning techniques to enhance loan profitability by optimizing the approval process and determining loan amounts based on risk assessment of applicants. It aims to accurately classify potential borrowers and predict the optimal loan amount for those approved. We faced challenges like class imbalance, which were addressed through a two-stage modeling approach, and we fine-tuned model performance using the F1 score and RMSE for classification and regression, respectively. The impact of the system is significant, allowing the bank to improve its lending decisions, thus increasing profitability while effectively managing credit risk.
Understanding Customer Attrition in Banking Code
Tech Stack: Python, Data Mining, Logistic Regression, SVC, KNN, GB, RF, Decision Tree, XGBoost
This project aims to mitigate customer attrition at XYZ Bank by developing a predictive model using machine learning techniques. The dataset comprises 10,000 entries with 14 columns, including key variables like CreditScore, Geography, Age, Balance, NumOfProducts, and Exited, the target variable indicating churn. Our workflow involves data partitioning, feature engineering, accuracy assessment, and model generalization to ensure robust predictions. This enables the bank to implement preemptive measures, minimizing churn and its adverse effects, ultimately improving customer retention and financial stability.
Stock Price Prediction and Anomaly Detection Code
Tech Stack: Python, GCP, Vertex AI, BigQuery, Cloud Pub/Sub, Dataflow, Data Studio
This project aims to leverage Google Cloud Platform (GCP) technologies to analyze historical and real-time stock data for predicting stock price movements and detecting anomalies. Using BigQuery, Cloud Pub/Sub, Dataflow, and Data Studio, and Vertex AI workbench we provide insights for informed investment decisions.
Key Features
- Real-time stock price prediction using historical and streamed data.
- Anomaly detection in stock price movements.
- Interactive dashboards for data visualization and analysis.
Translation of Gesture-Based Sign Language to Text and Speech Research Paper
Tech Stack: Python, CNN, NumPy, Image Classification, Supervised learning, Adam optimizer, ReLU
I conducted a study published in the Journal of Physics: Conference Series - IOPScience, where I employed supervised learning techniques to design convolutional neural network (CNN) architectures. These models exhibit remarkable efficacy in recognizing hand gestures, achieving an accuracy exceeding 96%. This advancement enables the seamless conversion of these gestures into both text and speech, thereby fostering improved communication channels for individuals with speech impairments, leveraging the power of image recognition technology.
Journal of Physics: Conference Series - IOPScience (doi:10.1088/1742-6596/1964/6/062074)
CERTIFICATIONS
