Manohar Eldhandi

• • Built a Django - based web application that helps determine if water is safe for drinking or irrigation, just by entering parameters like pH and turbidity.
• • Trained Random Forest and SVM models to classify water as potable or not, using real-world chemical datasets — the applicaton predicts instantly whether the water is drinkable.
• • Designed a modular backend system using Python, enabling seamless scaling and adaptability for batch processing via CSV or real-time streaming from IoT sensors.
• • Integrated model evaluation tools to show prediction accuracy and confusion matrix results, making it easy to trust what the model is saying.
• • Worked with a team of 3 — I handled backend + ML while others took care of frontend and documentation. We made sure it all came together like a finished product.

• • Built a command-line job portal backend using Java and MySQL, supporting user registration, login, job posting, and applications.
• • Designed REST-style APIs with raw Java and JDBC, enabling secure and efficient database operations without frameworks.
• • Implemented role-based access, encrypted credentials, and session tokens to ensure secure interaction for both user roles.
• • Normalized MySQL schemas for users, jobs, and applications; tested APIs rigorously using Postman for all major scenarios.
• • Structured code into `api`, `dao`, `model`, and `util` packages, ensuring modularity and future scalability.

• • Built a Java-MySQL inventory management system with full CRUD support for Products, Suppliers, and Orders via JDBC.
• • Designed normalized schemas and SQL scripts to maintain data integrity, quick lookups, and relational consistency.
• • Implemented modular architecture (`model`, `dao`, `service`, `utils`) to ensure code clarity and long-term scalability.
• • Enabled features like stock tracking, supplier management, sales reports, and product search for real business use cases.
• • Included smooth setup with connector configs and schema scripts — making deployment quick and developer-friendly.

• • Built a Java-based console application to simulate hospital operations like patient registration, appointments, and record management — all without using a database.
• • Designed core entities (`Patient`, `Doctor`, `Appointment`) using OOP principles such as encapsulation, inheritance, and polymorphism for clean, modular code.
• • Implemented in-memory and file-based data handling to mimic persistent storage, with features to add, update, and search patient and doctor information.
• • Delivered an intuitive command-line interface, ensuring smooth user interaction and efficient workflow simulation for hospital staff.

• • Developed a full-cycle data pipeline using Python to analyze customer transactions and demographics, enabling actionable segmentation.
• • Performed exploratory data analysis (EDA) and feature engineering to uncover meaningful patterns driving customer behavior.
• • Applied clustering algorithms like K-Means and DBSCAN, validating results with Davies-Bouldin Index and silhouette scores for optimal segmentation.
• • Built a Flask backend API delivering personalized product recommendations in real-time based on customer cluster profiles.
• • Showcased a seamless integration of data science and backend engineering to drive data-driven business intelligence and user personalization.

• • Developed a sentiment analysis model during my internship at Coding Raja Technologies, classifying Twitter text into positive, negative, and neutral sentiments.
• • Focused heavily on data preprocessing and feature engineering, optimizing the model for accuracy across diverse and noisy social media datasets.
• • Leveraged NLP techniques and machine learning with Scikit-Learn to extract meaningful insights from textual data for consumer behavior analysis.
• • Enabled organizations to make informed decisions by providing actionable market trend analysis based on real-time social media sentiment.

• • Developed a sentiment analysis model during my internship at Coding Raja Technologies, classifying Twitter text into positive, negative, and neutral sentiments.
• • Focused heavily on data preprocessing and feature engineering, optimizing the model for accuracy across diverse and noisy social media datasets.
• • Leveraged NLP techniques and machine learning with Scikit-Learn to extract meaningful insights from textual data for consumer behavior analysis.
• • Enabled organizations to make informed decisions by providing actionable market trend analysis based on real-time social media sentiment.

• • Mentored 100+ juniors in DSA under Smart Interviews, focusing on core concepts, coding patterns, and interview readiness.
• • Led doubt-clearing sessions, explained Java-based solutions, and simplified complex problems with real-world analogies.
• • Shared interview tips and coding strategies, helping peers boost confidence and tackle tech rounds effectively.
• • Enhanced my own skills in teaching, leadership, and structured problem solving through peer learning.

• • Selected among 3,000 out of 20,000+ applicants for Amazon’s exclusive ML Summer School program.
• • Completed intensive training in core machine learning topics including supervised/unsupervised learning, regression, decision trees, and model evaluation.
• • Built and fine-tuned models using Python libraries like scikit-learn and pandas on real-world datasets.
• • Strengthened understanding of ML math foundations — linear algebra, probability, calculus, and statistics.
• • Participated in problem-solving challenges and MCQs, reinforcing both theoretical concepts and practical coding skills.

Bachelor of Technology (B.Tech) in Computer Science & Engineering (AI & ML) at CMR Institute of Technology

• • Developed a water consumption forecasting system using XGBoost with 85% accuracy on real-world time-series sensor data.
• • Engineered features like hourly patterns, anomaly flags, and seasonal trends to improve model performance.
• • Preprocessed noisy data, trained multiple models, and deployed the final solution through a Python backend.
• • Showcased strong ML pipeline skills, rapid prototyping under pressure, and real-world environmental data handling.
• • Tools & Skills: XGBoost, Python, Time Series Forecasting, Data Cleaning, Feature Engineering.