Passionate about Cybersecurity & Artificial Intelligence Ar-Rasheed Smart University • Sana'a, Yemen
Building modern full-stack web applications, enterprise systems, and intelligent software with a focus on cybersecurity.
I am a Software Engineer and Full-Stack Web Developer passionate about architecting production web platforms, enterprise systems, and exploring cybersecurity and artificial intelligence.
My core work and focus areas:
- Enterprise & Security Systems: Architected and deployed an operations management platform for a private security company in Yemen (K9 Operations Management System)—featuring multi-tier Role-Based Access Control (RBAC), a production PostgreSQL database, multi-stage Docker containerization, and cryptographic security tokens protecting operational report approvals.
- Modern Web Architecture: Building responsive, high-performance web applications using Next.js 16, React 19, TypeScript, Python (Django/Flask), and Tailwind CSS with full bilingual (Arabic RTL & English) support.
- Cybersecurity & Hardware Defense: Researched hardware-level side-channel mitigations in confidential Processing-in-Memory architectures, accepted at IEEE IDEASET-2026 (se-pim-data-cleansing).
- AI & Multimodal LLM Defense: Engineered an upstream machine learning firewall protecting Vision-Language Models against typographic prompt injections, evaluated on Meta's official CyberSecEval 3 benchmark with 100% precision (visual_prompt_injection_detection).
- Systems Programming & Runtimes: Leveraging Rust (WebAssembly) and C++ when web applications demand high-speed constraint solving (opentimetable), container virtualization (CosmicReDroid), or low-level performance.
|
Modern Academic Timetable Studio combining a responsive Next.js 16 / React 19 web frontend with a high-speed native Rust constraint solver compiled to WebAssembly (WASM). Features a cleanroom reverse-engineered parser for aSc Timetables ( |
Accepted research paper and artifact evaluation for IEEE IDEASET-2026. Extends the SE-PIM confidential computing architecture with an in-memory data sanitization engine using fixed-schedule execution to prevent memory side-channel leakage. Verified through cycle-accurate gem5 simulations and strict reproducibility test harnesses. |
|
Upstream multimodal machine learning security firewall defending Vision-Language Models against typographic camouflage and visual prompt injection attacks. Achieves 100% Precision and 97.4% ROC-AUC on Meta's official |
Native Linux Android multi-instance control center built with Rust on System76's libcosmic desktop toolkit. Manages isolated ReDroid containers via Docker Bollard, featuring real-time hardware identity spoofing (Galaxy S24 Ultra, Pixel 8 Pro), deterministic BinderFS node isolation, DRI GPU acceleration, and zstd CoW snapshots. |
|
Operating systems scheduling simulator in C++ implementing 6 classical algorithms (FCFS, SJF, SRTF, Preemptive/Non-preemptive Priority, Round Robin) with scratch-built heap and queue data structures. Features real-time animated Gantt charts via Dear ImGui & OpenGL, backed by a 45-page formal engineering specification. |
Applied AI browser extension built on Chrome Manifest V3 that injects a context-aware conversational AI companion directly into web chat platforms (WhatsApp Web). Features dynamic DOM message extraction, sliding session memory, and asynchronous multi-LLM routing with zero telemetry leakage. |
Comprehensive operations platform engineered for a private security company in Yemen.
- Context & Confidentiality: Developed under client agreement for private security operations in Yemen (source code kept private; architecture and technical highlights shared with permission).
- Operational Scope: Managed the full lifecycle of security canine units—including tactical deployment schedules, handler duty assignments, veterinary medical logs, and breeding pedigrees.
- Security & Access Control: Multi-tier Role-Based Access Control (RBAC) isolating command leadership, veterinary staff, and field handlers.
- Cryptographic Token Verification: Implemented cryptographic security tokens to protect operational report generation and approval workflows against tampering.
- Architecture: Modular Python & Flask backend, production PostgreSQL relational database, and multi-stage Docker containerization.
| Domain | Technologies & Frameworks |
|---|---|
| Full-Stack & Web | • • • • • • • • • |
| Cybersecurity & AI | • • • • • • • |
| Systems & Architecture | • • • • • • • • |
| Methodologies & Tools | • • • • |