Skip to content
View danaaa913's full-sized avatar

Block or report danaaa913

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.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. 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
danaaa913/README.md

Dana Boarat

AI/ML Developer · Accessibility & Arabic NLP · Jordan

I build practical AI tools for underserved communities — focusing on Arabic speakers, visually impaired users, and students who lack access to quality resources. Most of my work is competition-driven, deployed on real devices or live backends.


Featured Projects

Project What it does Stack
Arshedni Campus navigation & safety app for visually impaired students at Yarmouk University. YOLOv8 object detection, KMeans clustering, SQLite data warehouse. Originally Flutter → migrated to native Android/Kotlin for real-time inference. Kotlin, YOLOv8, SQLite, Python
NovaCortex Arabic RAG-based Telegram chatbot for Yarmouk IT Faculty. 410+ knowledge-base entries, semantic cache, Graduation Simulator, Skill Gap Detector, Academic SOS. Built under NOVA AI team. Python, RAG, Telegram API
CPA — Cyber Protection Assistant Arabic digital-blackmail detection app submitted to the Crown Prince Foundation competition for best government app in Jordan. FastAPI backend deployed on Render with a live HTML dashboard. FastAPI, Python, Render
على قدي (Ala Addi) AI nutrition specialist mobile app for affordable local food alternatives — focused on Jordanian/Arab families and infant nutrition. Python, Mobile
OAODD Outdoor object-detection dataset: 2,988 YOLO label files across 9 classes, CC BY 4.0. YOLO, Computer Vision
Python Code Assistant Semantic retrieval with LangGraph routing, Groq inference, and a Gradio interface. LangGraph, Groq, Gradio
Toxic Comment Classification BiLSTM experiments for Arabic/multilingual toxicity detection with preprocessing and classification diagnostics. PyTorch, NLP

More experiments

PDF Context Retrieval · CodeLlama RAG · RAG Evaluation · Trash Instance Segmentation · Video Classification · Dental CNN · Multispectral Segmentation


Tools I use

AI/ML: Python · PyTorch · TensorFlow/Keras · YOLOv8 · scikit-learn · OpenCV · LangChain · LangGraph · ChromaDB · SentenceTransformers · AraBERT

Applications: Kotlin · Flutter · Java · FastAPI · Flask · React · TypeScript · Express · PostgreSQL · Gradio · Streamlit

Other: Unity/C# · SQLite · Render · Telegram Bot API


About

  • 🎓 AI student @ Yarmouk University, Faculty of Information Technology — Jordan
  • 🤝 Founder of NOVA AI, a student AI team
  • 🏆 Projects submitted to: Crown Prince Foundation (Jordan), university-level competitions
  • 🔍 Focus: Arabic NLP, accessibility tools, resource-constrained deployment

LinkedIn

Pinned Loading

  1. OAODD OAODD Public

    Outdoor object-detection annotations: 2,988 YOLO label files across 9 classes, with previews, citation and CC BY 4.0 licensing.

    2

  2. CodeGenerationExplanationChatbotUsingLangGraph CodeGenerationExplanationChatbotUsingLangGraph Public

    RAG retrieval and evaluation experiment using SentenceTransformers, ChromaDB, DialoGPT and DeepEval.

    Python 1

  3. Conversational-Agent-Using-LangChain Conversational-Agent-Using-LangChain Public

    PDF passage-retrieval prototype with LangChain, ChromaDB, LangGraph, FastAPI and Gradio conversation-history views.

    Jupyter Notebook 1

  4. DA-370-Data-Mining DA-370-Data-Mining Public

    Campus safety app for visually impaired students — YOLOv8, KMeans, SQLite, Android/Kotlin

    HTML 1

  5. DR3 DR3 Public

    RedLight AI prototype: feedback collection and administrative dashboards with Express, MongoDB and an external ML API integration.

    HTML 1

  6. TrashDetector TrashDetector Public

    Trash detection and segmentation using Detectron2 / Mask R-CNN

    Jupyter Notebook 1