AI-Assisted Python Code Review for Beginners
Applied AI Project | Python, Flask, Generative AI
Built a web-based tool using Generative AI to provide beginner-friendly Python code feedback. Implemented a Flask backend, HTML/CSS frontend, and structured editor interface to analyze logic errors, readability issues, best practices, and actionable improvement suggestions.
Integrated Gemini with prompt engineering, tested common beginner mistakes, and compared feedback quality across AI models for clarity, correctness, and usability.
Pokémon World Simulation
Advanced Programming Techniques | C, C++
Developed a Pokémon-style world simulation across ten progressive assignments, starting in C and later transitioning to C++ for object-oriented design, procedural terrain generation, multi-map navigation, trainer movement, and Pokémon encounters.
Implemented Dijkstra pathfinding, ncurses controls, CSV data parsing, object-oriented systems, inventory management, PokéBucks economy, and party/storage features for a cohesive interactive game loop.
Agentic AI Satisfaction Predictor
Customer Support Analysis & Data Exploration
Analyzed support ticket data to explore patterns in customer satisfaction (Low, Medium, High) using features such as response time, resolution time, agent experience, and reopened status. Cleaned the dataset, handled duplicates, and examined distributions to identify key factors for modeling.
Found strong class imbalance, with Low satisfaction dominating key segments across priorities, issue categories, and channels, helping guide future predictive modeling efforts.
Animal Image Classifier (CNN + Keras)
Image Recognition & Neural Networks
Built a CNN to classify Dogs, Cats, and Horses from image data using Keras. Preprocessed and normalized the dataset, trained the model, evaluated prediction performance, and analyzed misclassifications to better understand model behavior.
Demonstrated skills in deep learning, image preprocessing, and neural network evaluation, applying analytical methods to build and assess image classification models.
Archived Message Reconstruction
Data Structures in Java
Built a Java-based program to reconstruct archived message sequences using graph traversal algorithms like DFS and BFS on both directed and undirected graphs. Focused on correctly modeling relationships between messages to restore the original order.
Tested the implementation using trace-based methods to verify accuracy, while also comparing traversal approaches to understand performance differences and ensure efficient runtime behavior.
Travel Guide
Construction of User Interfaces
Built a multi-page responsive catalog website featuring product cards, FAQs, and author profiles using HTML, CSS, JavaScript, and Bootstrap. Focused on creating a clean, consistent interface with reusable components to ensure scalability and ease of maintenance.
Organized the project using a structured GitLab workflow, allowing for efficient development, version control, and collaboration while maintaining a stable and well-documented codebase.
github.com/anishnukala/Cy-Collectibles
