Automated spreadsheet input using Python and Google API services, reducing data entry errors by 95% and
saving 12+ hours/month in administrative workload.
• Developed a WhatsApp messaging bot using Python, improving staff communication efficiency, increasing
response rates by 40%, and reducing missed updates by 30%.
• Analyzed user engagement and operational data using Looker Studio, identifying trends that led to a 15%
improvement in program participation rates.
• Analyzed user engagement and operational data using Looker Studio, one-way ANOVA, t-tests, and regression
models (linear, logistic); visualized distributions using histograms and boxplots, leading to a 15% increase in
program participation rates.

Engineered an AI-powered platform from scratch that transforms user prompts into fully functional and styled
websites, reducing manual development effort by 80% for early users.
• Constructed dynamic front-end generation modules using React and TailwindCSS, automating component
creation and improving UI design consistency by 70%.
• Trained and fine-tuned transformer-based models on 10,000+ prompt-to-website mappings, boosting semantic
accuracy by 40%.
• Spearheaded backend integration using Node.js and FastAPI, reducing response latency by 50% with
caching and asynchronous task queues.

• Developed a real-time face recognition system using OpenCV and Python, automating attendance tracking
for 20+ students, reducing manual effort by 50%.
• Implemented facial detection and recognition algorithms with dlib and face recognition, achieving 80-85%
accuracy in face identification under controlled conditions.
• Engineered an attendance logging system, integrating Python pandas to organize and store attendance data
in structured formats (CSV/Excel), tracking 90% of daily attendance without errors.

Developed a credit card fraud detection model using machine learning algorithms in R, identifying
fraudulent transactions with an accuracy of 80-85%.
• Pre-processed and cleaned a large transaction dataset, handling missing values, normalization, and feature
extraction using dplyr and tidyverse packages.
• Identified key patterns and insights from transaction data using ggplot2, helping to identify trends and
anomalies in the data set.

• Developed an NBA Game Winner Predictor using Python and machine learning to forecast match
outcomes (win, loss, draw) with an accuracy of 75-80% based on historical data.
• Preprocessed and cleaned historical match data, including team statistics, player performance, and match
results, using pandas and NumPy to prepare the dataset for model training.
• Implemented machine learning models, including Logistic Regression, Random Forest, and XGBoost,
to predict match outcomes, achieving a precision of 70% for correct win/loss predictions.