Skip to content
View farwa-shaikh's full-sized avatar
6 followers · 0 following

Achievements

Achievement: Pull SharkAchievement: YOLOAchievement: Quickdraw

Achievements

Achievement: Pull SharkAchievement: YOLOAchievement: Quickdraw

Block or report farwa-shaikh

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
farwa-shaikh/README.md



🖥️  ~/about-me

role:           AI Engineer & Research Assistant / IT Manager
location:       Karachi, Pakistan
education:      B.E. Electronic Engineering, MUET
current_focus:  Adversarial AI · Deepfake Protection · LLM Agents
mission:        Making AI security research accessible to everyone
looking_for:    Masters/GRA opportunities · Professor collaborations
status:         actively building & researching

🔬  Research Focus


  • 🛡️ Building Patterned Shield — a browser-based tool that disrupts deepfake facial-landmark detection using Moiré patterns, asymmetric lighting & reflective hotspot simulation
  • 🧠 Researching GNNs for community detection, link prediction & co-authorship network analysis
  • 🤖 Building LangChain/LangGraph agents that explain why an image may be manipulated
  • 🌾 Co-authored an IoT paper on sugarcane disease detection (YOLOv8 + ResNet, ~98% accuracy)
  • ⚡ Implemented Louvain community detection + Isolation Forest from scratch in JavaScript — no Python backend, fully browser-based

🛠️  Tech Arsenal

AI / Machine Learning



Frontend / Creative Engineering



Security / Infra

🏅  Certifications

📌  Featured Project

🛡️ Patterned Shield

Browser-based deepfake image protection tool



A client-side tool implementing three adversarial disruption techniques — Moiré mesh overlays, asymmetric dual-lighting, and reflective hotspot simulation — to confuse facial-landmark detection used by deepfake AI. Zero server. Zero data upload. Runs entirely in-browser.



JavaScript  Canvas API  Adversarial AI  Computer Vision

Pinned Loading

  1. VoxAI-A-Dual-Mode-Virtual-Assistant VoxAI-A-Dual-Mode-Virtual-Assistant Public

    This project is a voice-controlled virtual assistant built in Python. It listens to your voice commands, recognizes speech

    Python 5

  2. Heart-Disease-Prediction-Using-Machine-Learning Heart-Disease-Prediction-Using-Machine-Learning Public

    Used supervised machine-learning algorithms

    Jupyter Notebook 6

  3. Finger-Counter-using-MediaPipe-and-OpenCV Finger-Counter-using-MediaPipe-and-OpenCV Public

    Run the code GameHandTracking.py only

    Python 1

  4. Drowsiness-Detector-using-Computer-Vision Drowsiness-Detector-using-Computer-Vision Public

    Install the packages, run the code, resolve the errors. Enjoy!

    Python 1

  5. Object-detection-using-yolo Object-detection-using-yolo Public

    I have used yolov8 for object detection, this project is first step of my main.

    Python

  6. Human-Activity-Recognition-using-Smartphone-Sensors Human-Activity-Recognition-using-Smartphone-Sensors Public

    I trained a deep learning model on smartphone accelerometer and gyroscope data to classify six human activities. The model achieved 97.5% validation accuracy and can be deployed in fitness tracking…

    Python