Kenechukwu Nneji
Data Science Intern @Velera
Signup · Get unlimited contacts
WORK HISTORY
Data Science Intern @Velera
FL, US
Designed and implemented an anomaly detection pipeline using Prophet, identifying unexpected record volume behaviors across timeseries data with a 30% improvement in detection accuracyTuned Prophet model performance through grid search on 64 hyperparameter combinations, reducing MAPE from 14.4% to as low as2.88% for seasonal job patternsEngineered temporal and contextual features (e.g, day_of_week, job_name, changepoint/holiday flags), improving model sensitivity toknown business cycles and and created volatility indicators using rolling std/Z-score to increase anomaly detection accuracyClustered jobs using KMeans and PCA based on autocorrelation and seasonality strength to generalize modeling across job groups withsimilar time series behavior (seasonal and non-seasonal patterns)
EDUCATION
Florida International University
Bachelor of Science - BS, Mathematical Data Science
ABOUT KENECHUKWU NNEJI
Data Science & AI major at FIU with strong foundations in statistical analysis, machine learning, and data analytics. Proficient in Python, SQL, Java, R, and JavaScript, with hands-on experience in predictive modeling, anomaly detection, data visualization, and cloud-based analytics. Skilled in tools like Scikit-learn, Prophet, Power BI, TensorFlow, and Databricks. Actively seeking roles across data science, machine learning, and analytics where I can apply statistical modeling and problem-solving.
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.