// RAG Engineer
RAG engineer for document analysis, retrieval and structured AI outputs.
My RAG work focuses on business documents where confidentiality, traceability and structured outputs matter. I care about retrieval quality, reviewability and how the AI result becomes useful software.
Best fit for teams with documents, reports, assessments or knowledge bases that need reliable AI-assisted analysis.
Proof points
- Building a RAG pipeline that analyzes EFQM self-assessment reports, detects conflicts and produces structured assessor-style outputs.
- Built a local-inference LLM routing system for prompt analysis and model selection.
- Designed the project around local LLM execution to protect sensitive organizational data.
- Connects retrieval, scoring methodology and report generation instead of treating RAG as a simple chatbot layer.
Relevant work
- May 2026 to present
EFQM/RADAR AI Report Analyzer
A local-first, multi-agent RAG pipeline for the first-pass desk assessment of EFQM-based organizational self-assessment reports. It retrieves assessment guidance, links findings to source evidence, flags gaps and conflicts, and produces structured draft feedback for human review. I lead the five-person team and enforce fail-closed scoring, typed agent contracts, reproducible run provenance, and CI evaluation gates.
Read case study - 2025 — 2026
LLM Routing System
An internal on-device LLM routing layer that classifies each prompt by task type and complexity before selecting a local model or cloud provider. My operational estimate was that roughly 65 percent of requests stayed local, reducing cloud AI spend by about 45 percent. These are self-reported estimates rather than figures from a controlled finance study.
Read case study - 2025 to 2026
Confidential AI/Technical Evaluation
Researched and benchmarked more than ten LLM approaches, AI tools, and hardware options for an early-stage internal evaluation. I tested selected options, documented evidence and uncertainty, and prepared decision-focused reports. The subject, candidates, findings, and outcome remain under NDA.
Read case study - Oct 2024 — Mar 2025
EUNO: ML-Based Mood Tracker (Graduation Project)
ML-based mood tracker with a Random Forest model that reached about 87 percent accuracy after feature engineering and SMOTE class balancing. A FastAPI backend runs the preprocessing and prediction path, while Flutter, SQLite, and Firebase support the application. Built with Jaser Quteshat under Dr. Ahmad Barghash.
Read case study
Search fit
- RAG engineer
- LLM engineer
- document AI engineer
- local LLM developer