I'm Ahmed Gaiter, an AI Engineer from Egypt focused on one thing: building AI systems that work in production, not just in demos.

My background is in backend engineering, and that shapes how I approach every AI project — with a bias for reliability, clean architecture, and systems that don't break when real users show up. I've worked across RAG pipelines, LLM integrations, multi-agent workflows, and safety evaluation — always with production deployment as the end goal.

Most AI engineers can run a Jupyter notebook. I build the thing that runs in production.

Not just basic vector search. I architect full RAG pipelines with smart chunking, embedding strategies, query rewriting, re-ranking layers, and source-grounded answers. Hands-on with both Pinecone and ChromaDB — I know when to use each.

I build FastAPI services that wrap LLM logic with proper validation, async design, error handling, and clean separation of concerns. APIs you can actually put behind a product.

Multi-step agent systems using LangGraph for stateful orchestration — tool use, memory management, and conditional branching. I think in workflows, not just prompts.

RLHF work at Outlier, evaluation at Alignerr, and Tier-3 auditing at Atlas Capture. I treat hallucination detection, guardrails, and evaluation loops as core parts of the product — not optional extras.

AI Engineer — RAG & Applied AI (Freelance · Jun 2023 – Present)
Building production-style AI applications independently — RAG systems, LLM backends, and automation pipelines.
Designed and shipped applied AI products using Python, FastAPI, LangChain, FAISS, Docker, Groq, and Next.js.
Built a multilingual RAG workspace (Arabic & English) with source-grounded answers and a chunk inspector layer.
Built a credit card fraud detection API using XGBoost with containerized FastAPI endpoints — 0.95+ ROC-AUC in testing.

AI Video Data Specialist & Quality Auditor — Tier 3 (Atlas Capture · Dec 2025 – Present · Remote)
Tier-3 auditor for Large Vision Models — data integrity, motion tracking, and annotation consistency across complex video datasets.
Performed temporal segmentation and labeling on video sequences to improve motion recognition quality with 95%+ precision.
Applied advanced labeling standards to upgrade training data quality and reduce pipeline errors in vision model systems.

AI Backend Engineer (Springer Capital · Internship · May 2025 – Sep 2025 · US Remote)
Built Python automation scripts and FastAPI services to support high-speed AI workflows in a trading/finance environment.
Designed containerized services with Docker and improved deployment reliability across Agile sprints.
Collaborated with remote engineers to integrate AI-driven features into backend workflows with CI/CD best practices.

AI Model Alignment Specialist — RLHF (Outlier · Part-time · Dec 2024 – Apr 2025 · Remote)
Evaluated and improved LLM code-generation quality (Python / SQL) through RLHF-style review and red-teaming.
Audited model outputs for correctness, security issues, and instruction adherence across coding and reasoning tasks.
Designed adversarial test cases to surface reasoning failures and unsafe behaviors in production LLM systems.

AI Reviewer (Alignerr · Part-time · Mar 2025 – Apr 2025 · Remote)
Evaluated LLM outputs for quality, correctness, and safety — contributing evaluation signals to model improvement pipelines.

IT Specialist (Hamza Emam Company · Full-time · Aug 2023 – Aug 2024 · On-site)
Technical support, network maintenance, and troubleshooting across hardware, software, and user-facing systems.

If you're building an AI product and need reliable engineering behind it — let's talk.

Ahmed Gaiter · AI Engineer · Egypt
ahmed-ai-dev · ahmedgeeter · portfolio