Mohammed Hossain

Motivated and results-driven professional pursuing a degree in Commerce and Computer Science at UNSW, seeking a position where I can leverage my expertise in information systems, advanced communication, collaborative teamwork, and innovative technologies to drive impactful solutions and contribute effectively to an organization’s success.

Minesweeper Game Project
Personal Software Development Project, C Programming

• Developed a command-line Minesweeper game in C, featuring user interaction with game commands, mine placement, and a dynamic hint system limited to three hints per game.
• Implemented gameplay functionalities including row and square mine detection, cell visibility toggling, and robust win/loss condition checks to enhance game dynamics and user experience.
• Utilized defensive programming to ensure error handling and game integrity, managing user inputs and game state transitions effectively.

Pokedex Management System
Software Development Project, C Programming

• Engineered a dynamic Pokedex application in C, enabling efficient storage, retrieval, and manipulation of Pokémon data using linked lists, encapsulating complex data structures and memory management.
• Implemented advanced functionalities such as adding, removing, finding, and displaying Pokémon with conditional filters based on discovery status and types, enhancing user interactivity and system robustness.
• Developed user-driven navigation and management features, including the ability to navigate through Pokémon entries, set discovery statuses, and manage Pokémon details based on user interactions, promoting an engaging user experience.

CS Beats Music Composition System
Software Development Project, C Programming

• Engineered a comprehensive music composition system in C, implementing over 50 interactive commands for real-time editing, navigation, and playback of musical compositions, enhancing user engagement and control.
• Developed advanced functionalities for adding and modifying musical notes, successfully managing validations across 12 musical keys and 10 octaves, which increased input accuracy and reduced error rates by 90%.
• Incorporated robust navigation and playback state management, enabling users to seamlessly move through tracks and toggle between playing and stopped states, significantly improving user experience and system responsiveness.
• Optimized memory management for dynamic data structures, achieving a 30% improvement in system efficiency by customizing garbage collection routines for a linked-list-based data management system handling hundreds of musical beats.

Text Processing and Command Line Utility
Software Development Project, Python Programming 

• Developed a Python-based command-line utility that interprets and executes simplified text manipulation commands similar to sed, enhancing functionality for batch processing of text files.
• Implemented comprehensive text manipulation features, including custom delimiter substitution, line-specific operations, and conditional text replacement, processing over 1,000 lines per second for efficient data handling.
• Designed robust error handling and user feedback mechanisms, ensuring clear communication of command errors and usage guidelines to prevent operational disruptions and streamline user experience.
• Integrated advanced pattern matching capabilities using regular expressions, allowing dynamic text searching and substitution based on user-defined patterns, which improved command flexibility and utility for diverse use cases.

UNSW Streams Communication Platform
Backend Development, Group Project

• Co-developed the backend for 'UNSW Streams', a digital collaboration and communication platform modeled after Microsoft Teams, to support UNSW's transition to hexamesters by facilitating enhanced group interaction and educational collaboration.
• Implemented key functionalities such as channel and server management, user authentication, direct messaging, server invitations, and comprehensive user roles and permissions, ensuring secure and flexible user interactions.
• Engineered advanced features including error handling mechanisms, user search capabilities, and administrative tools for user management within servers, mirroring complex functionalities found in commercial platforms like Discord and Teams.
• Collaborated with a team to design and integrate APIs for real-time data handling and user interactions, contributing to a robust and scalable server architecture that supports a high-intensity learning environment.

Content-Based Movie Recommendation System
Recommender System, Machine Learning, Python, Jupyter Lab

• Developed a content-based movie recommendation engine using Python and sklearn in Jupyter Lab, handling a dataset of 2,000 movies across 8 genres for a hypothetical streaming service.
• Implemented and benchmarked various machine learning models from the sklearn library, including Bernoulli and Multinomial Naive Bayes, achieving up to a 25% improvement in classification accuracy through optimized text preprocessing and feature selection.
• Engineered user simulation tests to evaluate system performance, defining and analyzing user profiles with tailored metrics that predicted user preferences with over 80% accuracy based on hypothetical user interaction data.
• Conducted a user study to gather qualitative feedback, utilizing this data to refine the recommendation logic over a simulated four-week period, thereby enhancing the overall system’s effectiveness.

BookBuddy: Hybrid Book Recommender System
Recommender System, Machine Learning, Python, Jupyter Lab

• Engineered 'BookBuddy', a hybrid recommender system in Python using sklearn in Jupyter Lab, processing a dataset of 200,000 books across 40 genres with both context-based and collaborative filtering techniques.
• Conducted comprehensive data preprocessing and utilized advanced similarity measures such as cosine similarity and Euclidean distance from sklearn, optimizing book matching accuracy.
• Achieved a precision accuracy of 80% in user testing, demonstrating the system’s capability to deliver relevant and accurate recommendations through precision and recall metrics enhanced by machine learning algorithms.
• Designed and validated user scenarios for different types of users, ensuring robust system performance and user satisfaction in a dynamic environment, with adjustments based on feedback from real user interactions.