Bridging Innovation & Code with AI & Full-Stack
Hi, I'm Alishba Mazhar. A Computer Science student at UET Lahore passionate about designing intelligent solutions using Machine Learning, Large Language Models, MERN Stack, and Flutter.
Professional Summary
I am a final-year Computer Science student at the University of Engineering and Technology (UET), Lahore, set to graduate in 2026. My passion lies at the intersection of traditional software engineering and artificial intelligence.
Throughout my academic journey, I have developed expertise across diverse stacks including the MERN (MongoDB, Express, React, Node.js) framework, .NET Core, Flutter, and Python. I specialize in creating AI-powered systems—ranging from automated version control utilities using Large Language Models to diagnostic machine learning architectures in healthcare.
As a researcher and developer, I constantly seek out challenging engineering problems. My academic achievements, coupled with hands-on development and teaching assistant roles, reflect my dedication to creating reliable, modern, and user-centric software.
Core Focus Areas
Generative AI & LLMs
Developing semantic RAG architectures, multi-agent frameworks, and prompt engineering utilities via LangChain.
Machine Learning
Supervised clinical diagnostic modeling, advanced class-balancing (SMOTE), and classifier benchmarking (Random Forest).
Flutter Mobile App
Developing highly responsive, cross-platform Android & iOS applications using Dart, state management, and REST API integrations.
Full-Stack Web Apps
Architecting secure enterprise MERN stack web portals, Express microservices, and high-performance ASP.NET Core backend Web APIs.
Skills & Technologies
Work Experience
Teaching Assistant (TA)
UET Lahore, Computer Science DepartmentDirecting and mentoring OOP (Object-Oriented Programming) laboratory sessions using **C++** and **C#** for junior students.
- Supervised students in understanding and executing object-oriented concepts: Encapsulation, Inheritance, Polymorphism, and Abstraction.
- Guided debugging sessions, designed coding assignments, conducted evaluations, and provided qualitative feedback to optimize logic building and algorithmic thinking.
- Developed auxiliary laboratory manuals and review materials to bridge academic lectures with hands-on implementation in integrated development environments.
Featured Projects
AutoGit
An AI-powered VS Code extension automating commit creation and mapping code evolution semantically.
SmartPouch
AI-powered expense tracking application that automates budget categorization and financial insights.
Eduvance
AI-powered SaaS school management platform streamlining timetables, LMS plans, and administrative workflows.
Nextay App
Complete hotel management platform integrating room booking, staff workflows, and a cross-platform Flutter UI.
Asthma Prediction Across Lifespan
Published research paper utilizing Machine Learning and Explainable AI (XAI) to predict asthma risk across patient lifespans with 95.15% accuracy.
Facial Attendance System
Attendance system using OpenCV face recognition and Flutter mobile dashboards to eliminate check-in bottlenecks.
Web Scraping & Sorting
High-performance data extraction scraping 100,000+ Etsy products, sorted and queried with custom algorithm implementations.
Research & Publications
Asthma Prediction Across Lifespan Using Machine Learning and Explainable AI
Core Achievements & Methodologies
Developed and published a high-accuracy machine learning framework diagnosing asthma severity by evaluating demographic matrices, multi-tiered medical profiles, and regional environmental indicators.
- Pre-processing Engine: Managed dense feature skewness and extreme class imbalances using SMOTE (Synthetic Minority Over-sampling Technique) to ensure model generalization across low-sample data groups.
- Model Architecture: Synthesized, evaluated, and tuned multi-class models (SVMs, Naive Bayes, Decision Trees) with the Random Forest model achieving peak accuracy levels of 95.15%.
- Explainable AI (XAI): Deployed model interpretability frameworks (SHAP/LIME) to map critical clinical decision pathways, highlighting key risk features (air quality, hereditary history) for healthcare providers.
- Metrics Optimization: Evaluated models using rigorous precision, recall, and F1-score matrices, aiming to minimize critical false-negatives in clinical risk estimation.
Education & Certifications
BS (Hons) Computer Science
UET Lahore
NLP and its Transformers
Coursera Certification
SQL (Basic, Intermediate & Advanced)
HackerRank Certification
Certificate of Appreciation
Higher Education Commission (HEC), Pakistan