[WIP]Staten Island OTP Forecasting
End-to-end machine learning project using MTA open data to forecast on-time performance, compare models, and explain predictions using SHAP.
I’m Chinmay Naringrekar, a data analyst at MTA focused on transit reliability, forecasting, dashboards, and machine learning projects that connect analytics with real-world operations.
Selected work combining real-world data, forecasting, recommendation systems, generative AI, and explainable analytics.
End-to-end machine learning project using MTA open data to forecast on-time performance, compare models, and explain predictions using SHAP.
Formula 1 race analytics and machine learning project using FastF1 to analyze Monaco 2025 race pace, tyre degradation, and lap-time prediction.
Personalized recommendation and analytics app built from drama metadata and personal ratings to analyze taste profile and suggest new shows.
Generative AI app that lets users ask questions about uploaded PDFs using document chunking, embeddings, FAISS similarity search, and LLM responses.
Machine learning assignment using cell nuclei features to train, test, evaluate, and visualize a decision tree classification model.
Planned healthcare analytics project focused on disease progression, multimorbidity patterns, risk prediction, and clinical decision-support storytelling.
business intelligence and visualization projects from my earlier portfolio, including Tableau and Excel dashboard work.
Tableau dashboard analyzing Seattle Airbnb listings to understand rental patterns, pricing factors, and strategy for optimizing short-term rental decisions.
Interactive Tableau dashboard exploring Netflix titles by type, country, rating, genre, release year, and detailed movie/TV show information.
Dashboard analyzing Spotify top tracks using song attributes such as key, acousticness, instrumentalness, danceability, valence, energy, and streams.
Tableau dashboard built after SQL exploration of global COVID-19 data, showing death percentage, infection rates, continent-level death counts, and projected trends.
SQL exploration of global COVID-19 data, showing death percentage, infection rates, continent-level death counts, and projected trends.
Excel dashboard using cleaned customer data to identify factors influencing bike purchases, with interactive filters and business-focused visual summaries.
A mix of analytics, machine learning, business intelligence, and operational reporting tools.
Excel, SQL, Python, pandas, exploratory analysis, KPI tracking, and operational reporting.
Regression, classification, model evaluation, forecasting, XGBoost, scikit-learn, and SHAP.
Power BI, Tableau, Streamlit, executive dashboards, monthly reports, and performance visuals.
SQL Server, Oracle, Trino, AWS S3, Redshift, Athena, and data pipeline organization.
My work focuses on converting messy operational data into clear insights, dashboards, forecasts, and decision-support tools. I’m especially interested in reliability analytics, predictive maintenance, and explainable machine learning.
I’m currently expanding my portfolio with projects that connect business intelligence, machine learning, and practical operational decision-making.