IdesignAIsystemsthatdeliver measurable results
Agentic systems, RAG pipelines, automation flows, and data processing solutions for forward-thinking teams.
20+ global clients — $200k+ in savings generated
About
The Engineer Behind the Systems
“Architecture-first thinking applied to AI systems. I design the full system overview and agent flow before writing a single line of code — then orchestrate everything to work together simultaneously.”
AI Engineer with 3+ years of experience architecting enterprise multi-agent platforms, production RAG systems, and machine learning pipelines. Currently building an AI platform serving 10,000+ engineers at Dar Al-Handasah, orchestrating 9 LLMs across 4 providers.
Engineers Served
Global Clients
Savings Generated
LLMs Orchestrated
Services
AI Systems That Work For You
Every service is designed to deliver one thing — real business results. Not demos. Not prototypes. Production systems that run.
Agentic Systems
AI that doesn't wait for instructions. Autonomous agents that plan, decide, and act across your tools and workflows — handling complex multi-step tasks end-to-end on your behalf, around the clock.
RAG Systems
Turn your internal docs, databases, and knowledge base into an AI that actually knows your business. Every answer is grounded in your own data — no hallucinations, no guesswork, always accurate.
AI Automation Flows
Stop paying people to do robotic work. I build intelligent automation pipelines that handle your repetitive tasks — email triage, report generation, data entry — running 24/7 so your team focuses on what actually moves the business forward.
Data Pipelines & ETL
Your AI is only as good as the data feeding it. I design extraction, transformation, and loading pipelines that move, clean, and route your data reliably — turning raw sources into fuel for intelligent systems.
Experience
Where I've Built
From computational design to enterprise AI — each role pushed me deeper into building systems that work at scale.
Sep 2025
Present
AI Engineer
Sep 2025 – Present
- Architected askAIDA, a multi-agent orchestrator platform serving 10,000+ engineers across 5 disciplines, routing queries to specialized subagents via LangGraph with real-time SSE streaming
- Built 5 domain-specific agentic RAG agents with per-domain search strategies achieving 90%+ retrieval precision
- Engineered GPU-accelerated document processing pipeline reducing parsing failures from 30% to under 5% across 20,000+ engineering PDFs
- Integrated Model Context Protocol (MCP) for dynamic tool discovery enabling a zero-code plugin architecture
Dec 2024
Sep 2025
AI Engineer
Dec 2024 – Sep 2025
- Built Botvers, a full-stack WhatsApp AI SaaS platform handling 500+ concurrent connections with multi-tenant session management
- Developed a Building Code Agent combining vision AI + RAG + LLM reasoning achieving 90%+ accuracy on compliance queries
Jan 2023
Oct 2024
AEC Software Engineer
Jan 2023 – Oct 2024
- Led a team of 4+ developers building generative floor plan optimization — 85% reduction in design time, 40x faster than manual
- Built reusable .NET geometry library and custom Rhino/Grasshopper plugins reducing calculation time by 60%
Interested in working together?
Book a CallPortfolio
Selected Projects
A selection of AI engineering projects — from enterprise RAG systems to generative design algorithms.
MCP Integration — Dynamic Tool Discovery
Integrated the Model Context Protocol into a multi-agent system for dynamic tool discovery at runtime. External MCP servers expose capabilities automatically — no code changes or redeployment needed.
AIDA Orchestrator Agent
Intelligent multi-agent orchestrator for an enterprise AI platform serving 10,000+ engineers. Routes queries to specialized domain agents with real-time SSE streaming, persistent memory, and multi-provider LLM fallback chains.
Domain-Specific RAG Agent System
Suite of 5 specialised agentic RAG systems for engineering disciplines. Autonomous retrieval with hybrid search, per-domain vector stores, and citation-backed responses grounded in proprietary documents.
Engineering Document Processing Pipeline
Production document processing pipeline converting thousands of engineering PDFs into searchable vector embeddings. GPU-accelerated parsing with LLM-powered visual extraction reduced processing failures from 30% to under 5%.
Building Code Agent
Intelligent RAG agent for Egyptian building code compliance. Cut regulatory search time by 80% with 90%+ accuracy across 1,200+ pages in Arabic and English.
AI Monitoring Platform
Self-hosted Langfuse deployment for enterprise AI observability in an air-gapped environment. 6-service Docker stack with Azure AD SSO, full LLM tracing, and cost tracking — deployed through enterprise security workflows.
Solar Radiation Predictor
ML system predicting solar radiation on building surfaces as an alternative to energy simulation. XGBoost achieves R² = 0.999 on known locations and 96.2% accuracy on unseen cities — millisecond inference replacing minute-long simulations.
WhatsApp AI Bot Platform
Enterprise-grade WhatsApp automation platform with AI integration handling 500+ concurrent users. Reduced customer service costs by 70% and automated 350+ daily inquiries at 99.5% uptime.
Generative Floor Plan
Automated floor plan generation using genetic algorithms in .NET. 85% reduction in schematic design time, 30% space utilisation improvement, and 40x faster than manual drafting — integrated with Rhino/Grasshopper.
Skills
Technologies & Expertise
Rated from real-world usage across production systems — not tutorials.
Certifications
Stanford Machine Learning Specialization
Dec 2024
Get Started
Ready to build something intelligent?
I partner with teams to design and deploy AI systems that solve real business problems. Let’s discuss your next project.