PhD Researcher, University of Essex/AI Engineer/Explainable AI
I build AI systems that can explain themselves — currently interpretable fuzzy classifiers for living neural biocomputers at the University of Essex. Before that, three years shipping production ML at BT Group and Saudi Motorsport. Published at WCCI FUZZ-IEEE 2026.
Five projects where the problem was hard, the constraint was real, and the result was measured. Each one has a full write-up — problem, method, results, and what I would do differently.
A Fuzzy-Based Approach for Interpretable Spike Detection in Living Neural Biocomputers
An interpretable fuzzy rule-based system for detecting neural spikes in biological neural networks grown on Multi-Electrode Arrays. Unlike black-box deep models, the approach produces human-readable IF-THEN rules that reveal which electrodes and firing patterns are associated with spike events — achieving a 97.74% F1-score across six biologically diverse chips.
IEEE World Congress on Computational IntelligenceMaastricht, Netherlands2026
04 Repository index
Everything else
Sixteen more public repositories across generative AI, LLMs, agentic systems, computer vision and applied ML. Filter by year.
Real-time ML for race operations, processing 13 telemetry channels per lap — throttle, brake, tyre temperatures and G-forces.
Streaming anomaly detection and lap-time prediction
Tyre degradation modelling and predictive maintenance
Containerised with Docker and Kubernetes for race-day reliability
2025 — Present
PhD in Artificial Intelligence
University of Essex, UK
Supervised by Prof. Hani Hagras and Dr. Michael Barros. Research: explainable fuzzy classifiers for living neural biocomputers.
2023 — 2024
Industrial MSc in Artificial Intelligence
University of Essex, UK
2019 — 2024
BA in Computer & Communication Engineering
Alexandria University, Egypt
06 Toolkit
Stack
Research
Fuzzy rule systems (T1 & IT2)
Genetic algorithms
Explainable AI
MEA / biocomputing
Feature selection
Statistical methods
Machine learning
Deep learning
Generative AI
LLMs & RAG
Computer vision
NLP
Reinforcement learning
Frameworks
PyTorch
TensorFlow
LangChain / LangGraph
HuggingFace
Scikit-learn
FastAPI
Infrastructure
Docker
Kubernetes
AWS
MLflow
CI/CD
Linux
07 Get in touch
Let's talk
I am open to research collaborations, AI engineering roles, and industry partnerships — particularly where explainability is a requirement rather than a nice-to-have. If you are working on something where a model has to justify itself, I would like to hear about it.