Presentation
🚀 What I Do:
✅ AI Agents & Automation: I build custom AI agents using LangGraph and LangChain that automate complex business workflows, replacing fragile human-dependent operations with robust, self-correcting systems.
✅ RAG Systems (Retrieval-Augmented Generation): I develop intelligent document querying systems that allow businesses to ask questions about their PDFs, Word documents, and internal data with accurate citations and sources.
✅ Custom Software Development: I write clean, maintainable Python code with secure API integrations (FastAPI) and database management (PostgreSQL, MongoDB).
✅ Security-First Approach: I apply offensive security principles to protect AI systems from prompt injection and other vulnerabilities.
💡 My clients get:
- 100% custom solutions tailored to their unique needs, not generic templates.
- Full source code ownership with clear architecture diagrams.
- English or Arabic documentation for easy deployment.
- 3 days of technical support after delivery.
🛠️ Technologies I work with:
Python, LangGraph, LangChain, FastAPI, RAG, Vector Databases (ChromaDB), n8n, Docker, Linux, REST APIs, MCP Protocol, Postgresql, Claude Code, OpenCode, Cursor.
📌 I work with businesses of all sizes—from startups needing a quick MVP to enterprises requiring complex automation ecosystems.
🌍 I am fluent in Arabic and English. I can provide full Arabic documentation and support if needed.
Let's turn your idea into a working system. Contact me today!
Portfolio
Offered services



Background
Key achievements and projects:
- AutoMB: Built an AI-powered orchestration framework for seamless multi-agent coordination with production-ready pipeline management.
- AtxAgent: Developed an advanced open-source ReAct framework featuring 18 multimodal tools (Web search, OCR, Audio transcription) with local (Ollama) and cloud LLM integration.
- CogniRAG: Designed an enterprise-grade RAG system capable of ingesting PDFs and URLs, storing them in vector databases (ChromaDB), and answering questions with accurate source citations.
- SecVuln: Created a fully automated penetration testing pipeline following the OSCP methodology.
Technologies: Python, LangGraph, LangChain, FastAPI, ChromaDB, Docker, n8n, REST APIs, PostgreSQL.
Key achievements and projects:
- CogniRAG: Built an enterprise-grade RAG system capable of ingesting PDFs, Word documents, and URLs, storing them in vector databases (ChromaDB), and answering natural language questions with accurate source citations.
- Advanced Document Processing: Developed ingestion pipelines that extract, chunk, and embed text from multiple document formats while preserving logical structure and metadata.
- Multi-Model Support: Integrated both cloud-based (OpenAI, Groq, NVIDIA) and local (Ollama) LLM models to ensure data privacy and flexibility.
Value delivered to clients:
✅ Accurate answers with source citations (no hallucination)
✅ Fast vector search across thousands of pages
✅ Full data privacy with local deployment options
✅ Customizable chunking and retrieval strategies
Technologies: Python, LangChain, ChromaDB, FAISS, FastAPI, PostgreSQL, OpenAI API, Ollama, Docker.
Key achievements and projects:
- SecVuln: Built a fully automated penetration testing pipeline using LangGraph and LLMs, covering network reconnaissance, service enumeration, CVE analysis, and automated reporting.
- Cyber-Core: Developed a modular red teaming toolkit for security automation and offensive operations.
- Security-First AI: Integrated offensive security principles into AI agent development to protect against prompt injection, command injection, and other OWASP Top 10 vulnerabilities for LLM applications.
Key capabilities:
✅ Automated vulnerability scanning and CVE correlation
✅ Network reconnaissance and service enumeration
✅ Security code review and hardening
✅ OSCP methodology applied to AI systems
This unique combination of offensive security and AI engineering allows me to build systems that are not only intelligent but also secure by design.
Technologies: Python, Bash, Kali Linux, Burp Suite, LangGraph, LLMs, Linux Hardening, OSCP Methodology.

