I design and build AI systems that work in real conditions. My focus is Edge-AI, computer vision and machine-learning architecture: models that run offline, on local hardware, where the data actually is.
I co-founded Raa'i, an award-winning offline livestock health-monitoring platform, founded the ANOMALY AI community (50+ members), and have worked across four ML internships and trainings in 2026.
| Role | Co-Founder and Project Lead. I led the AI architecture design and implementation. |
| Vision | Multi-model computer-vision pipeline: YOLOv8 with BoT-SORT and Re-ID for real-time cattle behavior tracking. |
| Reports | Offline retrieval-augmented generation (RAG) pipeline with a local LLM that writes bilingual veterinary diagnostic reports for farmers and vets. |
| Result | 1st place, Damietta AI Hackathon, Agriculture Track (37 teams). 2nd place, Egypt Cities Hackathon, with a 75,000 EGP prize. |
| Code | Private repository. |
The pipeline reads left to right: camera frames, detection, per-animal tracking, behavior monitoring, report generation. All of it runs on local hardware with no cloud dependency.
| What | Industrial gear defect detection with two independent detectors and one explainable decision. |
| CNN | A convolutional network written from scratch in pure NumPy, with no deep-learning framework. |
| Signal | An FFT-based anomaly detector working in the frequency domain. |
| Fusion | Both detectors feed an Explainable AI dashboard. |
| Data | A 2,000+ image synthetic dataset generated in Blender. |
| Code | DecodeLabs-Project2-Inspection-System |
A compiler that translates Arabic linguistic roots and diacritics into memory semantics and emits native x86-64 (PE64) binaries directly. It is now entering its self-hosting phase. In progress since March 2026.
| Repository | What it is | Stack |
|---|---|---|
| DecodeLabs-Project2-Inspection-System | Multi-modal Edge-AI inspection system for industrial gear defect detection: NumPy CNN, FFT signal analysis, explainable AI. | Python, NumPy |
| Decentralized-Autonomous-Police-Fleet-DecodeLabs | A decentralized blockchain architecture to secure cyber-physical systems and autonomous robot fleets. | Solidity |
| ITI-Graduation-Project-house-price-prediction | End-to-end ML web app that predicts Indian house prices with a Random Forest pipeline. | FastAPI, React, TypeScript, scikit-learn |
| NTI_Graduation_Project | Final project of the NTI internship. | Jupyter Notebook |
| ALFA-Kit | Advanced automation and OCR tool. | Python |
| Faculty-Attendance-Tracker | Faculty attendance tracking application. | TypeScript |
| muslim-app-release | Free, ad-free Muslim app: adhkar, Quran, prayer times, adhan and notifications. | Release repo |
More repositories
- codealpha_tasks: CodeAlpha Python programming internship tasks.
- breast_cancer_dataset_streamlit_task: Streamlit task for the NTI internship.
- NTI_Iris_dataset_for_streamlit: Iris model deployed with Streamlit (NTI internship).
- alfarouq637.github.io: personal site repository.
- Batman-Register-page: HTML5 register page, built for fun.
Languages
Machine learning and vision
Apps and tooling
| Role | Organization | Since |
|---|---|---|
| Co-Founder and Project Lead | Raa'i | Jul 2026 |
| Founder and CEO, AI/Data Science Architect | ANOMALY Community: 50+ members across 10 technical tracks | May 2026 |
| Vice Head | Google Developer Groups on Campus | Aug 2026 |
| Machine Learning Trainee | Information Technology Institute (ITI), Suez Canal Branch | Jul 2026 |
| Machine Learning Intern | National Telecommunication Institute (NTI), remote | Jul 2026 |
| Machine Learning Intern | FlyRank AI, remote | Jun 2026 |
| Public Relations Volunteer | Youth Leads Youth (YLY), Ministry of Youth and Sports | Jul 2026 |
Certifications: Build with AI: Masr Edition (Google, 2026) · Artificial Intelligence Ambassadors Program (MCIT Egypt, 2026) · Robotics and Automation Internship (DecodeLabs, 2026)
Open to AI Architecture and Machine Learning Engineering opportunities. The fastest way to reach me is LinkedIn or email.
