I am Tony Livins, a postgraduate (MSc) in Applied Artificial Intelligence and Data Science from Solent University, United Kingdom, holding a BSc (Hons) in Computing. My work focuses on AI engineering, data science, multimodal deep learning, and explainable AI, integrating research and development to design transparent, high-impact intelligent systems for real-world applications.

My MSc research project, "LungGuard- A Multimodal Deep Learning System for Early Lung Cancer Detection", combined chest X-ray images, tabular patient data, and clinical text using early, late, and hybrid fusion strategies. The project applied SHAP and Grad-CAM frameworks to enhance model interpretability, contributing to the wider goal of responsible and explainable AI in healthcare.

Languages

• Python, Java, JavaScript, C, C++, HTML/CSS, Bash
• R - statistical analysis and data visualisation
• SQL - advanced joins, window functions, and data analytics
• Markdown - for technical documentation
• PlantUML - for system architecture and diagram generation

Databases

• MySQL, SQLite, PostgreSQL, MongoDB
• Neo4j - graph-based reasoning and relationship modelling
• Firebase Realtime Database – prototype app integration
• AWS S3 Buckets - scalable data storage for large datasets
• BigQuery - cloud-scale querying and analysis
• Elasticsearch - semantic search and text analytics

Libraries

• NumPy, Pandas, OpenCV, scikit-learn, Matplotlib, Seaborn, Plotly, Dash, Statsmodels
• Hugging Face Transformers, NLTK, SpaCy – natural language processing
• SHAP, LIME, Grad-CAM – explainable AI and model interpretability
• PyCaret, WEKA - AutoML and model evaluation
• XGBoost, LightGBM - ensemble learning and gradient boosting
• NetworkX - graph analytics and relationship visualisation
• Altair, Tableau, Power BI – interactive visualisation and analytics

Frameworks

• Flask, Django, Node.js, TensorFlow, PyTorch, Keras (Sequential and Functional API)
• Streamlit - deployed the LungGuard multimodal system
• FastAPI - RESTful API design and model integration
• Hugging Face Transformers Framework – fine-tuning biomedical NLP models
• scikit-multimodal - multimodal fusion pipelines
• Apache Spark MLlib - distributed analytics and machine learning
• Bootstrap, Tailwind CSS - responsive UI design
• PyTorch Lightning - model training and checkpoint management

Tools and Technologies

• Git, Docker, AWS, GCP, Heroku, Firebase, JIRA
• Google Colab, Jupyter, VS Code, PyCharm, Spyder
• Kaggle - dataset management and reproducibility
• MLflow - experiment tracking and benchmarking
• Streamlit Cloud, Hugging Face Spaces - cloud-based deployment
• Anaconda, Miniconda - environment management
• Tableau Desktop and Public – interactive dashboards
• GitHub and Git LFS - version control and model storage
• Trello and Notion - project and research tracking
• LaTeX and Overleaf - dissertation and technical documentation
• Linux CLI - scripting and virtual machine operations

Core Competencies

• Machine Learning and AI - supervised and unsupervised learning, multimodal deep learning, NLP, and explainable AI
• Data Visualisation and Analytics - Tableau, Power BI, and Python-based dashboards
• Cloud Deployment - AWS EC2, Dockerised microservices, Streamlit Cloud
• Software Engineering - modular coding, testing, version control, documentation
• Statistical Analysis - correlation, ANOVA, regression, and diagnostics
• Responsible AI - fairness, bias mitigation, and ethical design principles
• Research and Project Planning - Gantt charts, agile task tracking, and resource allocation
• Academic Communication - Harvard referencing, synthesis, and presentation

Specialist Expertise

• Multimodal Deep Learning (image, text, and tabular fusion)
• Explainable AI (SHAP, Grad-CAM)
• Clinical AI Ethics and Model Transparency
• Biomedical NLP (Bio_ClinicalBERT and domain adaptation)
• Deployment-ready AI Pipelines for healthcare applications
• Statistical Significance Testing (McNemar’s Test)

I aim to contribute to research and innovation that bridge data-driven intelligence, ethical AI, and human-centred design. My long-term goal is to drive progress in artificial intelligence through rigorous research, technical excellence, and measurable real-world impact.