I am a second year CSE Cybersecurity student with a keen interest in web development, cybersecurity and AI/ML

github

Yatin Annam  • Peerlist

This project involved building a stock price prediction tool using Python and machine learning. Historical stock data was fetched using the yfinance API and processed to predict the next day's closing price. It focused on making short-term forecasts using real-world financial data. A Linear Regression model was trained on the closing prices and evaluated using Mean Squared Error to assess prediction accuracy. The dataset was split into training and testing sets to validate performance. Visualization was done using matplotlib to compare actual vs predicted prices. The project helped me understand time series forecasting, regression modeling, and data preprocessing. It also strengthened my skills in model evaluation and working with structured datasets. This task demonstrated how AI can support decision-making in financial applications.

AI for Predicting Stock Prices

Built a Secure File Sharing System using Python and Flask. The project allows users to securely upload and download files through AES encryption (CBC mode), ensuring data privacy. It features a clean and basic web interface, proper file separation (uploads, encrypted, decrypted), and uses a .env file to manage encryption keys securely. This project helped reinforce my understanding of Flask, file handling, and cryptographic principles in real-world applications.

Secure File Sharing System

This project is a classic implementation of the Tic-Tac-Toe game using Python, where the player competes against an AI opponent. The game is played in the terminal, and users make their moves by selecting positions on a 3x3 board. The AI uses the Minimax algorithm to make optimal decisions, ensuring it either wins or forces a draw. The algorithm evaluates all possible outcomes recursively, choosing moves that minimize the player’s chances of winning. The project includes full game logic: input handling, board display, win/draw detection, and alternating turns between the player and the computer. It was built using core Python concepts like lists, loops, conditionals, and recursion.

Tic-Tac-Toe AI

Used Splunk SIEM to monitor and analyze simulated web access logs for potential security threats. Sample log files were ingested and queried to identify anomalies such as repeated timestamps and suspicious URI access patterns. Two incidents were documented: a possible reconnaissance attempt detected through clustered log entries with identical timestamps, and suspicious access behavior based on HTTP error codes and endpoint paths. The final report includes detailed search queries, incident classification, remediation suggestions, and screenshots from the Splunk dashboard.

Security Alert Monitoring & In

Simple chatbot that responds to user queries like product info, support issues, return policies, and general help. Uses basic Python concepts like string manipulation, conditional statements (if-elif-else), and loops to simulate a help desk assistant. The chatbot takes the user's name, detects keywords from their input, and gives appropriate responses. It runs in a loop until the user exits by typing something like “bye” or “exit.”

Rule-Based Chatbot

This project focused on identifying security vulnerabilities in the Damn Vulnerable Web Application (DVWA), a deliberately insecure web app used for learning and practice. It was hosted on Kali Linux using Docker for safe local testing. The vulnerabilities tested included SQL Injection, Reflected Cross-Site Scripting (XSS), Broken Authentication through brute-force login, and Cross-Site Request Forgery (CSRF). Each attack was simulated manually through the browser interface. All findings were mapped to the OWASP Top 10 categories. Each vulnerability was documented with screenshots, technical details, and suggested mitigation strategies to improve application security.

Web Application Security Testi

Built an end-to-end machine learning model to detect fraudulent credit card transactions using a real-world dataset containing over 284,000 records. The project involved cleaning the dataset, balancing the data using undersampling techniques to address class imbalance, and applying Logistic Regression to classify transactions as legitimate or fraudulent. Achieved an accuracy of ~92% with strong performance metrics (precision, recall, F1-score). The entire pipeline was implemented using Google Colab and Python libraries like pandas, NumPy, and scikit-learn. Tech Framework: Python, pandas, NumPy, scikit-learn, Google Colab, GitHub This project helped me understand the real-world challenges in fraud detection like class imbalance and model evaluation, while also improving my data preprocessing and model-building skills.

Credit Card Fraud Detection us

Weather Forecast Webpage: A basic weather forecast webpage that fetches real-time weather data using the Open-Meteo API. It allows users to search for any city and view temperature, humidity, wind speed, and weather conditions. Tech Framework: HTML, CSS, JavaScript, Open-Meteo API, GitHub Pages This was my first attempt at building a web-based project while learning how to work with APIs. Excited to enhance and expand it further!