🎓 B.Tech Data Science Student | 📊 Data Science & Analytics | 🤖 Machine Learning & AI | 🐍 Python Developer | 🌐 Django Developer

I build practical, data-driven applications that turn real-world problems into useful solutions. My work spans data analysis, machine learning, artificial intelligence, Generative AI, NLP, computer vision, and web application development using technologies such as Python, SQL, Pandas, Scikit-learn, Streamlit, Django, Ollama, and Qwen.

I'm passionate about learning by building, experimenting with new technologies, and developing end-to-end projects that combine data, AI, and software development.

🌐 Portfolio: https://rinkitala-commits.github.io/jhumarani-portfolio/

A local AI chatbot built with Streamlit, Ollama, and Qwen 2.5 3B that provides conversational AI responses while keeping the AI model running locally.

Tech Stack:

Python • Streamlit • Ollama • Qwen 2.5 3B • JSON

🔗 GitHub Repository

• 🤖 Local AI chatbot powered by Qwen 2.5 3B
• 🦙 Ollama local LLM integration
• 💬 Interactive Streamlit chat interface
• 🧠 Conversation memory
• 💾 JSON-based local chat persistence
• 🎚️ Adjustable response creativity
• 📊 User and AI response statistics
• 📥 Conversation download functionality
• 🔍 Ollama model availability checking
• ⚠️ AI response error handling
• 🗑️ New Chat functionality
• 🔒 Local-first AI architecture
• 🐙 Git/GitHub version control

Built a complete local AI chatbot application demonstrating LLM integration, conversational AI, Streamlit application development, session-state management, local data persistence, and practical AI application development.

An AI-powered resume analysis and job-matching application that analyzes resumes against job descriptions, identifies matching and missing skills, evaluates resume quality, and recommends suitable job roles.

Tech Stack:

Python • Pandas • NumPy • Scikit-learn • NLP • TF-IDF • PyPDF2 • python-docx • Streamlit

🔗 Live Demo

🔗 GitHub Repository

• 📄 PDF and DOCX resume parsing
• 🧠 Automated resume skill extraction
• 💼 Job description skill extraction
• 🎯 Resume-to-job matching
• 📊 TF-IDF based text similarity
• 🧩 Skill-based matching
• 📈 Combined job match scoring
• 📋 Resume quality analysis
• ✅ Matching skill identification
• ⚠️ Missing skill detection
• 🔥 Job-specific skill analysis
• 💡 Personalized resume improvement recommendations
• 💼 Suitable job-role recommendations
• 📊 Skill comparison visualization
• 📈 Match score and skill-match progress indicators
• 🖥️ Interactive Streamlit dashboard
• ☁️ Streamlit Community Cloud deployment
• 🐙 Git/GitHub version control

Built an end-to-end AI-powered resume analysis platform demonstrating NLP, TF-IDF text similarity, skill extraction, machine learning concepts, data processing, recommendation logic, and interactive Streamlit application development.

A machine learning web application that predicts whether a customer is likely to churn based on customer demographics, services, contract details, and billing information.

Tech Stack:

Python • Pandas • NumPy • Scikit-learn • Matplotlib • Seaborn • Joblib • Streamlit

🔗 Live Demo

🔗 GitHub Repository

• 🧹 Data cleaning and preprocessing
• 📊 Exploratory Data Analysis (EDA)
• 🔧 Feature engineering
• 🔢 Numerical feature scaling
• 🏷️ Categorical feature encoding
• 🤖 Logistic Regression classification
• 🌲 Random Forest classification
• ⚖️ Model comparison
• 📈 ROC-AUC analysis
• 🎯 Precision-Recall analysis
• 🧩 Confusion matrix evaluation
• ⚙️ Hyperparameter tuning using GridSearchCV
• 📊 Accuracy, Precision, Recall and F1-score evaluation
• 🎯 Churn probability prediction
• 🟢🟡🔴 Customer churn risk assessment
• 👤 Customer summary dashboard
• 🖥️ Interactive Streamlit application
• ☁️ Streamlit Community Cloud deployment
• 🐙 Git/GitHub version control

Built an end-to-end machine learning application that transforms customer data into churn predictions and probability-based risk assessments through an interactive web interface.

A machine learning regression web application that predicts house sale prices based on property characteristics such as overall quality, living area, location, year built, garage capacity, basement area, and other housing features.

Tech Stack:

Python • Pandas • NumPy • Scikit-learn • Matplotlib • Seaborn • Joblib • Streamlit

🔗 Live Demo

🔗 GitHub Repository

• 🧹 Data cleaning and preprocessing
• 📊 Exploratory Data Analysis (EDA)
• 🔍 Feature correlation analysis
• 🧩 Numerical and categorical feature handling
• 🔢 Numerical feature imputation and scaling
• 🏷️ Categorical feature encoding
• 🔄 Scikit-learn preprocessing pipelines
• 🌲 Random Forest Regressor
• 🚀 Gradient Boosting Regressor
• ⚖️ Regression model comparison
• 📈 MAE, RMSE and R² evaluation
• 📊 MAPE evaluation
• ⭐ Feature importance analysis
• 💾 Model serialization using Joblib
• 🏠 Interactive house price prediction
• 🎛️ User-controlled property features
• 🔄 Reset input functionality
• 📊 Prediction details and model metrics
• 🖥️ Interactive Streamlit application
• ☁️ Streamlit Community Cloud deployment
• 🐙 Git/GitHub version control

The Gradient Boosting Regressor was selected for the final application based on the evaluation results on unseen test data.

The trained model identified several influential features, including:

• Overall Quality
• Above Ground Living Area
• Garage Cars
• Total Basement Area
• First Floor Area
• Basement Finished Area
• Second Floor Area
• Year Built
• Lot Area
• Year Remodeled

Built an end-to-end machine learning regression application demonstrating data preprocessing, exploratory data analysis, feature engineering, categorical encoding, model comparison, regression evaluation, feature importance analysis, model serialization, and interactive Streamlit application development.

An end-to-end banking analytics project that analyzes transaction data, identifies fraud patterns, performs SQL-based business analysis, trains machine learning models for fraud detection, and presents insights through an interactive Streamlit dashboard.

Tech Stack:

Python • Pandas • NumPy • Scikit-learn • SQL • SQLite • Matplotlib • Seaborn • Streamlit • Joblib • Git • GitHub

🔗 Live Demo

🔗 GitHub Repository

• 🏦 Banking transaction data analysis
• 🧹 Data cleaning and preprocessing
• 🔍 Exploratory Data Analysis (EDA)
• 🗄️ SQLite database integration
• 🧮 SQL-based transaction analysis
• 📊 Advanced SQL queries and business insights
• 🔎 Fraud pattern analysis
• 💰 Transaction amount analysis
• 📅 Time-based transaction analysis
• 🌍 Location-based fraud analysis
• 📱 Transaction channel analysis
• ⚙️ Feature engineering
• 🤖 Logistic Regression fraud detection
• 🌲 Random Forest fraud detection
• ⚖️ Machine learning model comparison
• 📈 Accuracy, Precision, Recall, F1-score and ROC-AUC evaluation
• 🎯 Fraud probability prediction
• 📊 Interactive fraud analytics dashboard
• 🚨 High-risk transaction identification
• 📈 Monthly fraud trend analysis
• 🎛️ Interactive dashboard filtering
• ☁️ Streamlit Community Cloud deployment
• 🐙 Git/GitHub version control

Built an end-to-end banking analytics platform demonstrating SQL analysis, data cleaning, exploratory data analysis, feature engineering, machine learning, fraud detection, database integration, and interactive Streamlit dashboard development.

An AI-powered application that automatically analyzes videos using visual motion and audio energy to identify exciting moments and generate a highlight reel.

Tech Stack:

Python • OpenCV • Librosa • MoviePy • NumPy • Pandas • Streamlit

🔗 Live Demo

🔗 GitHub Repository

• 🎬 Video motion analysis
• 🔊 Audio-energy analysis
• 🧠 Excitement scoring
• 🎯 Automatic highlight selection
• 🎞️ Automated highlight reel generation
• 📊 CSV analysis exports
• 🖥️ Interactive Streamlit web application
• ☁️ Streamlit Cloud deployment
• 🐙 Git/GitHub version control

An end-to-end Data Science application that collects job listings from a publicly accessible job-data API, cleans and analyzes real-world job data, performs salary and skill analysis, and presents insights through an interactive Streamlit dashboard.

Tech Stack:

Python • Requests • Pandas • Matplotlib • REST API • Streamlit • CSV • Git • GitHub

🔗 Live Demo

🔗 GitHub Repository

• 🔎 Automated job data collection using a REST API
• 🧹 Real-world data cleaning and preprocessing
• 💰 Salary analysis and salary midpoint calculation
• ⚙️ Feature engineering
• 🧠 Skill-demand analysis
• 🌍 Job location analysis
• 🏢 Company listing analysis
• 🏠 Remote, Hybrid, and In-Office work-mode analysis
• 🔍 Job search functionality
• 🎛️ Interactive filtering
• 📊 Data visualizations
• 💡 Automated dashboard insights
• 🔗 Direct job application links
• 🖥️ Interactive Streamlit dashboard
• ☁️ Streamlit Cloud deployment
• 🐙 Git/GitHub version control

An interactive financial analytics application that helps users track income, expenses, savings, and overall financial health through data-driven dashboards and visualizations.

Tech Stack:

Python • Streamlit • Pandas • Matplotlib • SQLite • Git • GitHub

🔗 Live Demo

🔗 GitHub Repository

• 📊 Interactive financial KPI dashboard
• 💰 Income and expense tracking
• ➕ Add and 🗑️ delete transactions
• 🔎 Category and date-range filtering
• 📈 Income vs. expense analysis
• 💵 Savings rate and savings progress tracking
• ❤️ Financial health summary
• 📊 Spending distribution and top spending categories
• 📅 Monthly financial analysis
• 📥 CSV financial report downloads
• 🗄️ SQLite database integration
• 🧩 Modular Python project architecture
• ☁️ Streamlit Cloud deployment
• 🐙 Git/GitHub version control

A full-stack e-commerce web application built with Python and Django that provides a complete online shopping experience with product browsing, search, cart management, wishlist functionality, checkout, order tracking, stock management, and Django Admin management.

Tech Stack:

Python • Django • SQLite • HTML • CSS • Bootstrap • Django Templates

🔗 Live Demo

🔗 GitHub Repository

• 👤 User registration, login and logout
• 🛍️ Product listing and product detail pages
• 🔎 Product search functionality
• 🏷️ Category-based product filtering
• 💰 Price and name-based sorting
• 📦 Product stock availability tracking
• 🛒 Shopping cart functionality
• ➕ Increase and decrease cart quantities
• 🗑️ Remove products from cart
• ⚖️ Stock validation during cart operations
• ❤️ Wishlist functionality
• 💳 Checkout and order creation
• 📋 Order history
• 🚚 Order status tracking
• 📉 Automatic stock reduction after purchase
• 👨‍💼 Django Admin customization
• 📊 Product and order management through Admin
• 🔍 Admin product search and category filtering
• 🗄️ SQLite database integration
• 🧩 Django MVT architecture
• 🔗 Django ORM and database modeling
• 🔐 Authentication and session management
• ⚙️ CRUD operations
• 🐙 Git/GitHub version control

Built a complete Django-based e-commerce platform demonstrating backend development, database integration, authentication, product and inventory management, shopping-cart workflows, order processing, and admin-side management.

• Python
• SQL
• C

• NumPy
• Pandas
• Data Cleaning
• Data Analysis
• Data Visualization
• Exploratory Data Analysis (EDA)
• Statistical Analysis
• Feature Engineering

• Machine Learning
• Classification
• Regression
• Logistic Regression
• Random Forest
• Gradient Boosting
• Model Evaluation
• Artificial Intelligence
• Computer Vision
• Feature Analysis
• Predictive Analytics
• Data-driven Applications

• Generative AI
• Large Language Models (LLMs)
• Ollama
• Qwen 2.5
• Local LLM Integration
• Conversational AI
• Prompt Engineering
• AI Application Development

• Natural Language Processing (NLP)
• TF-IDF
• Cosine Similarity
• Text Similarity
• Skill Extraction
• Resume Parsing
• Keyword Matching
• Skill Gap Analysis
• Recommendation Systems

• Django
• Django MVT Architecture
• Django ORM
• Django Templates
• Django Admin
• HTML
• CSS
• Bootstrap
• REST APIs
• API Data Collection
• Streamlit
• Interactive Web Applications

• SQLite
• SQL
• Django ORM
• Database Modeling
• CRUD Operations

• Scikit-learn
• OpenCV
• Matplotlib
• Seaborn
• MoviePy
• Librosa
• SQLite
• CSV
• Joblib
• Git
• GitHub
• Streamlit Cloud
• Ollama

• 📊 Advanced Data Science
• 📈 Excel & Data Analysis
• 🤖 Machine Learning
• 🧠 Artificial Intelligence
• 💬 Generative AI & LLM Applications
• 👁️ Computer Vision
• 🐍 Advanced Python
• 🌐 Django & Web Development
• 🗃️ SQL
• 📈 Data Visualization
• 🔬 Statistics for Data Science
• ⚙️ Machine Learning Model Optimization

• Data Science
• Machine Learning
• Artificial Intelligence
• Generative AI
• Large Language Models
• Computer Vision
• Data Analytics
• Python Development
• Django Development
• Full-Stack Web Applications
• REST API Applications
• Predictive Analytics
• Building practical AI applications
• Building Machine Learning applications
• Turning real-world data into useful insights

I enjoy learning by building real-world projects and continuously improving them with new technologies.

I'm continuing to build practical projects in:

• 🤖 Machine Learning
• 🧠 Artificial Intelligence
• 💬 Generative AI & LLMs
• 📊 Data Science
• 📈 Data Analytics
• 👁️ Computer Vision
• 🌐 Django & Web Development
• 🔗 API-based Applications
• 🖥️ Interactive ML Applications

My goal is to turn real-world problems into useful, data-driven applications.

Feel free to explore my repositories and projects.

🔗 GitHub

🔗 LinkedIn

⭐ Thanks for visiting my profile!

💡 I learn by building, experimenting, and turning ideas into practical projects.