- Tools: Data Preparation, ML: random forest, XGboost, Scikitlearn
- Description: Build a machine learning model to predict which chemical compounds can fight HIV effectively, helping researchers focus on the most promising candidates and skip compounds that likely won't work. Eg: You have 40,000 chemical compounds to test. Instead of testing all in the lab (expensive and slow!), you use AI to predict which might work.
- Tools: Modeling - Decision Tree, Random Forest, and XGBoost, Correlation Analysis
- Description: Objective1 : is it possible to classify songs into genres with just audio features \n Objective2 : what can these audio features tell us about the distinctions between genre (Naive Bayes, Decision Tree, KNN)
- Tools: Python, NLP, scikit-learn, NLTK
- Description: Built sentiment classification models (Naive Bayes, Decision Tree, KNN) on customer feedback with accuracies of 78.66%–79.83%. Preprocessed text data using CountVectorizer, label encoding, and NLTK techniques. Created word clouds, sentiment distribution plots, and confusion matrices to visualize insights.
- Tools: Python, Machine Learning, Streamlit
- Description: Developed a web app using the Texas housing dataset (txhousing) to forecast sales trends. Enabled user-driven predictions for city-specific and global trends. Integrated EDA, model selection, and real-time output using a Streamlit-based UI.
- Tools: Shell, Airflow and Kafka
- Description: Extracted and integrated data from SQL queries, APIs, and web scraping using both ETL and ELT approaches. Designed scalable pipelines to feed a centralized data warehouse, improving data accuracy by 25%. Completed hands-on labs using Kafka and Airflow.
- Tools: Python, scikit-learn, Deep Learning
- Description: Predicted weekly sales for 45 Walmart stores using store-level sales data, holiday indicators, and economic factors. Performed data cleaning, feature engineering, and model tuning using Gradient Boosting, Decision Trees, and Deep Neural Networks, improving forecasting for inventory and promotion planning.
- Tools: TensorFlow, Deep Learning, Image Segmentation
- Description: Designed a U-Net based segmentation model to detect crop rows from aerial field images. Optimized performance using IoU (Intersection over Union) and improved image preprocessing and annotation strategies.
- Tools: R, Data Visualization, R shiny, Statistics
- Description: Explored county-level Midwest U.S. census data to identify patterns in income, education, and population distribution. Presented findings through statistical summaries and visualizations to uncover regional disparities.
- Tools: ML, AI, Chatbot, RAG, SQL, DBMS, DLSG softwares, Freeflow, Metadata, Batch Processing, ETL, Data warehouses, Delta lakehouse, Schema design.
- Tools: SQL, Azure, Apache Airflow, GitHub, Restful APIs, Flask, ETL/ELT, CI/CD pipelines, SQL, NoSQL, Data warehouses, Delta lakehouse, SAP ERP, Tableau, Power BI.
- Tools: Oracle DB, HP ALM, Python, Automation testing scripts, Data warehouses, Litmus Magix, Selenium.
-
Master of Science in Applied Data Science | Indiana University Indianapolis | Jan 2023 – May 2024 | Dean’s Scholarship Recipient
- Coursework: Data Analytics using Python and R, Data Visualization, Deep Learning, Cloud Computing, DBMS, Statistics
-
Bachelor of Engineering | Mechanical, Aeronautics | Mangalore Institute of Technology and Engineering, VTU, India
I'm open to collaborating on interesting projects or discussing new opportunities. Feel free to reach out!
- 📧 Email: [email protected]
- 🔗 LinkedIn: LinkedIn Profile
- 🔗 Portfolio: Portfolio
- 🔗 Github: Github
- 🔗 Rpubs: link
- 🔗🔗 Medium