BSc Software Engineering graduate (University of Stirling) building AI-powered systems — from agentic RAG pipelines to ML-driven mobile apps. Aspiring AI/Data Engineer based in the UAE.
- Lab Sample Intake Agent — A single-agent workflow that extracts structured fields from documents via a local LLM, validates them against business rules, and flags uncertain records for human review instead of silently trusting AI output.
- NASA AI Research Assistant — An agentic RAG system that answers questions from NASA research PDFs, with automatic fallback to live web search when the answer isn't in the documents.
- UAE Lead Gen AI — An AI pipeline that finds, scores, and enriches business leads for a UAE licensing firm, with confidence-flagged contact data and personalized outreach drafts.
- CarbonLife — My final year honours dissertation: a real-time sustainability app combining a Kafka/WebSocket streaming pipeline with dual ML + NILM recommendation systems (86% accuracy) to help UAE residents cut their carbon footprint.
Languages: Python, Java, JavaScript AI / LLM Engineering: RAG, Agentic Tool Use / Function Calling, Prompt Engineering, Vector Databases (ChromaDB), LLM APIs (Gemini, local models via Ollama), Structured Extraction & Validation Machine Learning: LightGBM, XGBoost, Random Forest, scikit-learn Data & Infrastructure: Apache Kafka, WebSockets, PostgreSQL, MongoDB, REST APIs Tools: Docker, Git, React Native, JWT, n8n
⭐️ Currently exploring: the theory behind LLMs, RAG, and agentic systems — going deeper past the "how to build it" into the "why it works."