SANJUKTHA DONE

Highly motivated and detail-oriented AI & Machine Learning enthusiast with a strong foundation in Artificial Intelligence, Natural Language Processing (NLP), and predictive modelling. Proficient in developing and optimizing machine learning models for text generation, summarization, and sales forecasting. I have in-depth knowledge in sentiment analysis, text summarization, and named entity recognition (NER). Skilled in implementing machine learning algorithms, including Random Forest, XGBoost, and Gradient Boosting, with expertise in feature engineering, hyperparameter tuning, and model evaluation. Programming skills in Python and some NLP libraries. Passionate about solving real-world problems through AI-driven solutions and eager to contribute to innovative projects in the field of Generative AI.

• Management & Finance Team, College Chapter, Organized 5+ tech events, handled budgeting & logistics.
• Event Coordinator, Managed technical workshops, driving 200+ student participation.

• Problem-solving and algorithmic thinking, with experience in coding challenges and optimization techniques.
• Enhanced data pre-processing pipeline for a machine learning library, resulting in a 2x improvement in model training speed and a 15% reduction in memory consumption for image processing tasks.
• Participated in hackathons to develop innovative solutions.

Project 1. Prediction of Sales of a New Product

• Built a machine learning-based forecasting system to predict new product sales using XGBoost, Gradient Boosting, Random Forest, and Linear Regression, enabling better inventory planning, marketing strategies, and sales forecasting.

• Performed model comparison and evaluation, using MSE, MAE, and R² scores—XGBoost delivered the highest accuracy among all models tested.

• Applied advanced feature engineering techniques, including TF-IDF on product descriptions, sentiment analysis on reviews, one-hot encoding, and feature scaling to enhance model performance.

• Developed a real-time interactive Flask web app for live predictions; tools used: Python, scikit-learn, XGBoost, Flask, pandas, and NumPy.

Project 2: Three-Level Password Authentication System

•  Developed a 3-level authentication system using Text Password, Colour Pattern, and OTP for enhanced security.

•  Built with PHP, HTML, JavaScript, MySQL; deployed via XAMPP with modular architecture.

•  Secured against SQL injection, bots, and brute-force; tested for 1000+ concurrent users.

•  Scalable and user-friendly, with session logging and future-ready for biometric integration.

• AWS Cloud Practitioner Essentials, Completed AWS Cloud Training covering core services, architecture, security, pricing models, and cloud migration strategies.
• AWS Data Analytics, Completed training on Amazon EMR covering EMR Serverless and Cluster deployments, use cases, cost structure, and hands-on tutorials for big data processing using Apache Spark, Hive, and Presto.