Ali Imtiaz
Full-Stack & Mobile Dev | React Native & Flask Specialist
New to Contra
Ali is building their profile!
Project Title: AuraScore — Your Social Vibe Rated
AuraScore is an AI-powered social profile analysis platform designed to turn the way you present yourself online into something fun, insightful, and shareable.
The idea is simple: upload a screenshot of a social media profile, and AuraScore uses AI vision to analyze the profile’s overall aesthetic, branding, content presentation, personality cues, and digital presence. It then transforms that analysis into a personalized Aura Score and a premium, shareable AuraCard.
Instead of giving users a generic rating, AuraScore provides a unique breakdown of their online persona, including their Aura Identity, personality traits, observations, funny quotes, improvement suggestions, and overall social vibe. When a recognizable public figure or creator is detected, AuraScore can also generate a playful, lighthearted roast based on their public online persona.
The platform is built around the idea of making AI analysis entertaining rather than purely technical. Every result is designed to be something users can screenshot, download, share with friends, or post on social media.
Key Features:
• AI-powered visual profile analysis
• Personalized Aura Score
• AI-generated Aura Identity and traits
• Profile observations and improvement suggestions
• Funny, personalized commentary
• Playful celebrity/creator roasts
• Instagram handle extraction directly from screenshots
• Premium dynamic AuraCards
• Downloadable and shareable results
• Guest and authenticated user experiences
• Responsive web experience
AuraScore was built with a strong focus on AI, modern UI/UX, social sharing, and viral product mechanics. The goal is to bridge the gap between useful AI analysis and the kind of fun, personalized experience people naturally want to share online. From a development perspective, the project explores AI vision, structured AI outputs, dynamic UI generation, authentication, responsive design, client-side image generation, social sharing, and cross-platform web compatibility.
1
6
Project Title: FilmGenie Full-Stack AI Movie Recommendation Platform
Overview: FilmGenie is a live, full-stack AI-powered movie discovery web platform designed to eliminate choice paralysis. It features an intelligent recommendation engine that analyzes user inputs—such as age, region, favorite genres, and personal preferences—to generate personalized film selections in real time through a dark-mode web interface.
Key Features & Technical Execution:
1: AI Recommendation Engine: Architected dynamic recommendation pipelines to process user taste profiles and deliver contextual movie suggestions.
2: RESTful Backend Architecture: Built a modular Node.js and Express backend handling user preference requests, dynamic filtering, and TMDB API integrations.
3: Database & Cloud Deployment: Configured MongoDB Atlas for data management and deployed the production-ready application to Vercel with optimized performance and low latency.
4: Responsive UI/UX: Designed a modern, dark-themed React frontend tailored for both mobile and desktop viewports.
Tech Stack:
Frontend: React, JavaScript, HTML5, CSS3
Backend: Node.js, Express.js
Database: MongoDB Atlas
Cloud & Deployment: Vercel
APIs: TMDB API, AI Integration
1
14
Project Title: AI Learning Hub Machine Learning & Generative AI Web Platform
Overview: AI Learning Hub is a unified web application designed to serve end-to-end Machine Learning, Deep Learning, and Generative AI inference pipelines. Built with a responsive Flask backend and modular Python data pipelines, the platform delivers real-time model execution, data preprocessing, and evaluation metrics across classical ML and advanced neural network architectures.
Key Features & Technical Execution:
1: Unified ML Interface: Integrated classical machine learning pipelines (Linear Regression, Naive Bayes) with deep learning architectures (CNNs, RNNs, LSTMs) using PyTorch and Scikit-learn.
2: Generative AI & NLP Integration: Leveraged Hugging Face Transformers to deliver real-time sentiment analysis, text generation, and voice-based Q&A features.
3: Full-Stack Integration: Architected a modular Python / Flask RESTful backend with interactive UI controls, allowing users to input raw data, trigger model inferences, and inspect evaluation metrics seamlessly.
4: Data Processing & Pipeline Optimization: Implemented custom preprocessing and feature extraction pipelines to handle multi-modal inputs (text, voice, and tabular data).
Tech Stack:
Language: Python
Backend Framework: Flask
Machine Learning & Deep Learning: PyTorch, Scikit-learn
Generative AI & NLP: Hugging Face Transformers
Data Processing: NumPy, Pandas
1
27