Rishab Ashok

Data science enthusiast with internship experience in data preprocessing, exploratory analysis, and machine learning model development. Proficient in Python, Power BI, SQL, and Excel. Committed to leveraging data-driven insights to address business challenges effectively.

• Python Basics – University of Michigan (2024-05 – 2024-07)
• Python for Data Science, AI & Development – IBM (2023-08 – 2023-11)

Netflix user behavior analysis :

Cleaned 8,500+ Netflix records using Python and engineered features for analysis

Visualized trends in genre, rating, and release timing using Seaborn and Matplotlib

Derived viewer insights, such as top genres and dominant countries of origin

https://github.com/rishabashok/netflix-data-analysis

Uber trip demand forecasting :

Forecasted daily Uber demand using Random Forest (MAPE: 7.14%) and XGBoost, and created time-based features (lags, rolling averages) for improved accuracy

Built a Power BI dashboard to visualize actual vs. predicted trips

https://github.com/rishabashok/uber-trip-demand-forecasting

Supply chain optimization and demand forecasting :

Predicted product demand using a random forest model and Streamlit dashboard, found revenue leaders, and cost-defect relationships through EDA 

Enabled CSV exports and filtering by product/location for real-time insights

https://github.com/rishabashok/supply-chain-analysis

Genomic disease prediction using machine learning :

Collected and cleaned raw genomic data from public repositories and patient records, applied normalization, and converted categorical traits into numerical formats to ensure consistency and compatibility with machine learning models.

AI-DRIVEN WOMEN SAFETY MODEL:

Captured and transcribed voice commands using the Whisper model, preprocessed text, and classified intent using Naive Bayes to trigger safety alerts and location sharing