.jpg)
Projects
Hackathon Projects - LabLab.ai
Falcon AI71 Hub Hackathon Video Github Certificate Mental Health Coach Technologies Used: Django, Next.Js, Onrender - Created an Artificial Intelligence system that utilizes a finetuned model in order to serve as a coach for mental health.
- Implemented real-time personality selection allowing users to choose their type and communicate with the coach.
Llama 3 Hackathon Video Github Certificate AI News Analyzer Technologies Used: Langchain, Llama 3, C/C++, Python - The tool allows users to upload an article, which it then recognizes and processes through sophisticated text recognition algorithms.
- We have developed two key agents within the AI News Analyzer: the News Editor and the News Analyst.
Final Year Project - Natural Language Processing & Computer vision
Translip - Real-time Translation & Lip Syncing Github Technologies used: OpenAI, Wav2Lip, Flask, Llama 2 Libraries: Keras, Scikit, Tiktoken, Pytorch, Tensorflow - Translated English videos into Urdu language for students. Utilized OpenAI Whisper and Google Translate.
- Implemented lip reanimation techniques to synchronize video presenters’ lip movements with translated Urdu audio, using Wav2Lip and LipGan.
- Applied NLP with Llama 2 to generate personalized explanations for technical terms in video content.
- Bridged language gaps by integrating translation, lip reanimation, and personalization for Urdu-speaking students.
Item Menu Detection - Computer vision
Colab Report - Technologies used: Yolov8, Roboflow, Google Colab
- Managed and augmented datasets on Roboflow.
- Trained Yolov8 model on the data for menu item detection.
- Trained the model for 75 Epochs on Nvidia V100 GPU.
- Detected menu items with high confidence values above 80%.
CICD Pipelines using Azure DevOps - Cloud Computing
Technologies used: Github, Azure DevOps Github - Source Code & CI: Managed code using version control system with CI pipeline and unit/automated tests.
- CD & Release: Set up a CD pipeline with release gates and deployment rings (slow, fast, production).
- Infrastructure & Containers: Used Infrastructure as Code (IaC) and containerization for scalable deployment.
- Monitoring & Collaboration: Implemented using Azure DevOps for monitoring and Slack for instant notifications from dashboards.
Diabetes Prediction Using Random Forest Classifier
Technologies used: Pandas, Streamlit, Seaborn, Scikit-Learn, Plotly Github - The application provides an intuitive and interactive interface where users can input relevant medical data such as glucose levels, blood pressure, BMI, and other health indicators to generate a report on their diabetes risk.
- The application uses visual graphs to compare the user’s data with the dataset, providing an educational tool to help users better understand their health status.
- Users receive an easy-to-understand report indicating whether they are diabetic, coupled with visual feedback.
Movie Recommender System Using Knowledge Based Recommendation
Technologies used: NumPy, Pandas, Scikit-Learn, and NLTK Github - This project introduces a movie recommendation system using a content-based approach. The system is designed to recommend movies similar to a given movie based on various attributes.
- Approach focusing on calculating cosine similarity between movies’ vectors. These vectors are derived from attributes such as overview, genres, keywords, cast, and crew members, ensuring a comprehensive analysis for each recommendation.
- Streamlit library is utilized to create an intuitive frontend, allowing users to easily interact with the system and discover movie recommendations.
More Projects
FoodCart using pure C language implementation. Github Subnet-Calculator using python tkinter for visualization. Github Inventory-Management-System. SQL-Implementation using PHPmyadmin. Github SuperCart implementing OOP Principles in C++. Github Design & Analysis of Algorithms - Graham Scan, Jarvis March, Quick Hull Github