Javad Jomehpour Chahar Aman

End-to-End ML Pipeline Development: Designed and implemented scalable data pipelines using Azure and AWS, integrating CI/CD workflows to streamline data science projects from ideation to production.

• Predictive Modeling and Forecasting: Developed advanced time series models to forecast revenue and financial metrics, enabling data-driven decision-making and improving forecasting accuracy.
• Architectural Design and Deployment: Led the creation of robust, cloud-based architectures for machine learning projects, ensuring reliability, scalability, and seamless integration with business operations.
• Generative AI Initiatives: Pioneered the adoption of Generative AI within the finance department, helping the team leverage advanced AI capabilities to drive innovation and efficiency in financial forecasting and analysis.

• Big Data Analytics: Conducted data analysis on a large dataset of smartphone gps data using Python to detect movement patterns of Dallas, TX residents post-COVID-19, resulting in the identification and visualization of significant behavior changes due to the pandemic
• Classification and Clustering: Utilized logistic regression and XGBoost models and clustering models (k-means and DBSCAN) to investigate nonlinear relationships of built environment attributes and users’ mobility patterns, leading to prediction of the trip destination of users with a high degree of accuracy
• Deep Learning: Scraped over 60,000 tweets and App reviews related to E-scooters and employed topic modeling and CNN architecture with TensorFlow and Keras to characterize public opinion toward services, uncovering key topics and sentiments around E-scooter services
• Leadership: Served as the vice president of the Graduate Students Council of Lyle Engineering Departments, leading and managing a team of students to execute initiatives and projects related to the engineering department

• Communications: Collaborated with a team of six engineers and communicated with clients to identify the projects’ scope, needs, and deadlines and ensure their satisfaction with the outcomes
• Machine Learning: Developed and tested a predictive model (Random Forest) for traffic speed limit data to ease data collection efforts
• Interactive Dashboard: Analyzed Federal Highway Administration's (FHWA) travel monitoring outputs (over 35 million row datasets) and utilized interactive dashboards and visualizations (Tableau and Power BI) to reveal spatio-temporal patterns in trucks traffic volume