I am an AI Engineer based in Amman, Jordan, with experience developing RAG applications, evaluating LLM reliability, and building reproducible coding-agent benchmarks for international clients.
My benchmark work includes writing task instructions, reference solutions, deterministic verifiers, automated tests, and adversarial edge cases in Docker-based environments. I have also held reviewer and adjudication responsibilities, assessing agent reasoning, tool use, and output correctness.
Previously at Menadevs, I developed AI applications and Python workflows using LLMs, embeddings, vector search, FastAPI, Pinecone, and ChromaDB.
My portfolio includes a Hybrid RAG Knowledge Assistant combining BM25 and semantic retrieval with source citations, and an LLM Evaluation & Red-Team Lab with English and Arabic test cases and reproducible reports.
I am seeking remote freelance, contract, or part-time opportunities in AI evaluation, coding-agent benchmarks, RAG development, and Python backend engineering. I am available for 15–20 hours per week.
