Data & Impact

Neelima Verma (She/Her)

Jersey City, NJ ********@****.*** +1-551-***-**** LinkedIn: Neelima Verma - Data Science GitHub: vneelima44 PROFILE

Ex-Customer Service Head transitioning into a Data Scientist role in FinTech after completing a Master’s in Data Science from Pace University, focusing on data-driven financial models and machine learning applications. Diverse experience leading customer service teams, leveraging data for customer insights, and optimizing financial processes in the FinTech space. Over 10 years of experience collaborating with C-suite executives, driving digital transformation, and optimizing business processes that enhanced customer experience and operational efficiency. Developed predictive models and advanced analytics solutions that resulted in a 30% reduction in customer onboarding time and improved decision-making for financial services. PROFESSIONAL EXPERIENCE

Reel Works – Data & Impact Intern (Extended through Fall 2025) New York, NY May 2025 – Present

• Enhanced data accuracy and reliability by identifying and correcting Salesforce data pipeline errors, improving reporting trust for leadership and board meetings.

• Strengthened survey workflows by reviewing and optimizing Typeform survey data, building automated validation checks, and improving feedback collection processes.

• Leveraged Salesforce Tooling API to document all standard/custom objects, metadata, and lookup relationships, building a central reference to enhance reporting accuracy and system transparency.

• Built a Python script with the Typeform API to extract survey metadata (titles, types, options, required status), streamlining intake form audits. Reviewed Listen4Good surveys and co-designed Pre/Post & Anonymous instruments to improve data quality.

• Developed Tableau dashboards for sign-in trends and alumni engagement; validated data and prepared metrics/visuals for Q3 Board Meeting reporting. Proposed dashboard enhancements and created Loom tutorials to improve staff adoption.

• Designed and bulk-uploaded data templates for nonprofit/school partnerships; updated historical after-school and alumni data; designed FY26 departmental layouts in Canva for Salesforce dashboard translation. Home First Finance Company India Limited Feb 2014- Aug 2024 Team Lead & Head of Customer Service Mumbai, India Sept 2017- Aug 2024

• Led predictive analytics initiatives, designing a propensity model using logistic regression to forecast a 20% churn rate, enabling targeted retention strategies and reducing early attrition.

• Developed an Excel-based automated branch assessment tool, identifying performance gaps and increasing customer satisfaction by 25%.

• Trained 150+ customer service managers, driving operational improvements and higher service standards across regions.

• Implemented a case management system that improved turnaround time (TAT) by 40%, streamlining issue resolution and enhancing service delivery.

• Conducted data-driven service audits of 100,000+ customer accounts, identifying gaps and creating targeted training programs to address key issues.

Home First Finance Company India Limited Gujarat, India Area Customer Service & Operations Head Sept 2016-Aug 2017

• Led operational optimization for 5+ branches, enhancing service performance through strategic planning and customized training programs.

• Honored with the prestigious CSM Legend Award at the national level for securing the highest loan signings and achieving 100% insurance penetration, while maintaining an NPS score above 80 for financed customers. Home First Finance Company India Limited Gujarat, India Customer ServiceOperations & Relationship manager (Sales) Feb 2014-Aug 2016

• Spearheaded the setup and growth of HomeFirst's first branch in Surat, driving the branch to become the top-grossing location within the first year, surpassing revenue targets by 25%.

• Achieved 100% insurance penetration by cross-selling insurance with home loans, driving revenue growth.

• Closed 7 loans in 7 days while managing a high-risk collection portfolio, achieving the 7-Up Challenge. TECHNICAL SKILLS

Programming Languages: Python, SQL, PostgreSQL

Statistical Methods: Statistical Modeling, Regression Analysis, Hypothesis Testing Machine Learning Algorithms: Linear Regression, Logistic Regression, Random Forest, XGBoost, K-Means Clustering, Decision Trees, Support Vector Machines (SVM), KNN, Natural Language Processing (NLP) Data Manipulation: Pandas, NumPy, Data Preprocessing, Feature Engineering, Data Cleaning, Excel Libraries: Scikit-learn, TensorFlow, PyTorch, Matplotlib, Seaborn Data Visualization Tools: Tableau, Power BI, Matplotlib, Seaborn Database Management: PostgreSQL, MySQL

ACADEMIC PROJECTS / PERSONAL PROJECTS

Customer Segmentation Using K-Means Clustering Jan 2025

• Applied K-Means clustering to segment customers based on purchasing behavior, enabling targeted marketing strategies for improved customer engagement.

• Conducted data preprocessing, including outlier handling and feature scaling, to ensure accurate model performance and meaningful segmentation.

• Segmented customers using the RFM (Recency, Frequency, Monetary) model and visualized results through box plots, elbow curves, and cluster snapshots to assess the optimal number of clusters. Recommendation System Using User-Based Collaborative Filtering Jan 2025

• Built a recommendation system using collaborative filtering to suggest movies based on user ratings, enhancing user experience by personalizing content recommendations.

• Preprocessed the MovieLens dataset by filtering out low-rated movies, improving the quality and accuracy of the recommendation results.

• Created a user-item matrix to compute user similarities using Pearson correlation and Cosine similarity, ensuring relevant suggestions.

• Visualized the recommendation process and results using Python libraries (Pandas, Seaborn) to demonstrate the effectiveness of the system in enhancing user engagement. Dynamic Pricing Using Machine Learning Jan 2025

• Developed a machine learning model for dynamic pricing to optimize product prices based on real-time demand, seasonality, and customer behavior.

• Preprocessed data by standardizing numerical features using StandardScaler and encoding categorical variables using One- Hot Encoding.

• Applied Random Forest Regressor to predict optimal prices, capturing complex relationships between features.

• Evaluated model performance with Mean Squared Error (MSE) and Mean Absolute Percentage Error (MAPE) to ensure accuracy and reliability.

• Conducted feature importance analysis to identify key drivers affecting price prediction, such as demand, seasonality, and customer behavior.

Churn Prediction using Random Forest, Decision Tree and XGBoost Feb 2025

• Built a churn prediction model to identify at-risk customers and enhance retention strategies.

• Preprocessed data using Label Encoding for categorical features to maintain model efficiency and interpretability.

• Addressed class imbalance using SMOTE (Synthetic Minority Oversampling Technique), improving prediction reliability for the minority class (churned customers).

• Trained and evaluated Decision Tree, Random Forest, and XGBoost models with 5-fold cross-validation, achieving the best performance with Random Forest.

• Assessed model performance beyond accuracy, focusing on precision, recall, and F1-score to evaluate the model's effectiveness in identifying at-risk customers.

Credit Card Fraud Detection using machine learning Feb 2025

• Trained Logistic Regression, Random Forest, and Decision Tree models on imbalanced data, focusing on Precision, Recall, and F1 Score to assess fraud detection performance beyond accuracy.

• Implemented under-sampling and over-sampling techniques to address class imbalance, ensuring a more balanced and accurate dataset for model training.

• Incorporated Scikit-learn for building machine learning models and Seaborn/Matplotlib for visualizing model performance and comparison of evaluation metrics.

• Analyzed and compared model results before and after sampling, demonstrating significant improvements in fraud detection metrics (Precision, Recall, F1 Score).

Sales Prediction using XGBoost Feb 2025

• Developed a sales prediction model using XGBoost, handling missing data with interpolation and feature engineering to improve performance.

• Compared Random Forest and XGBoost, with XGBoost outperforming in accuracy (higher R score).

• Conducted exploratory data analysis (EDA) and cross-validation to assess model robustness. Sentiment Analysis Using NLP Feb 2025

• Performed sentiment analysis on customer reviews in the FinTech domain using Natural Language Processing (NLP) to classify reviews as positive or negative.

• Preprocessed raw text data by removing special characters, tokenizing, lemmatizing, and eliminating stopwords to enhance model performance.

• Implemented CountVectorizer and TF-IDF Vectorizer to convert text data into numerical format for machine learning models.

• Built and evaluated different machine learning models including Random Forest, XGBoost, and Decision Tree classifiers to predict sentiment.

• Achieved competitive accuracy and model performance by analyzing evaluation metrics such as accuracy and confusion matrix.

EDUCATION

Pace University - Seidenberg School of Computer Science and Information Systems New York Master of Science (MS) in Data Science Concentration: Data Science May 2026 IBS Hyderabad Hyderabad, Andhra Pradesh

MBA in Maketing with HR Concentration: Marketing Feb-2014 ADDITIONAL TRAINING & CERTIFICATIONS

Accenture North America, Data Analytics and Visualization Job Simulation 2024

• Analyzed 7 datasets to identify trends and provide actionable insights, simulating the role of a Data Analyst in a real-world business context.

Citi Finance, Finance Job Simulation 2024

• Created reports and presentations on financial risk analysis, summarizing key data points for senior management. Quantium, Data Analytics Job Simulation 2024

• Performed data analysis to deliver commercial insights using Python, enhancing strategic decision-making. Lean Six Sigma Black Belt, KPMG & Green Belt, Benchmark Six Sigma 2024

• Lean Six Sigma Black Belt (KPMG) and Green Belt (Benchmark Six Sigma) certifications, providing expertise in process optimization, statistical analysis, hypothesis testing, regression analysis, root cause analysis, and implementing data-driven strategies to drive efficiency, reduce costs, and improve quality in business processes. ACTIVITIES

Data Science Club, Pace University

Outreach Team Leader Sep 2024 – Present

• Connected with 50+ industry experts via LinkedIn and cold emailing, organizing 5+ events and seminars through collaboration.

• Monitored +7 events and initiatives, engaging 150+ students with the data science community and expanding the club's network by 40% through professional connections.

Contact this candidate