I help organisations turn AI and digital capability into measurable business outcomes, bridging the gap between what technology can do and what the business actually needs.

Over the years, I have worked as a Product Owner and Business Analyst across enterprise AI platforms, conversational AI systems, and data-driven digital products, predominantly in complex, cross-functional environments where stakeholder alignment and clear requirements were the difference between delivery and drift.

My work spans the full delivery lifecycle: from eliciting requirements and mapping processes in discovery workshops, through to writing user stories, defining acceptance criteria, managing backlogs, and monitoring product performance post-launch. I have hands-on experience evaluating Generative AI and LLM capabilities against real business problems, and translating that analysis into roadmap decisions stakeholders can act on.

Alongside industry delivery, I research the governance and regulatory dimensions of AI adoption, with a specific focus on EU AI Act compliance and GDPR in enterprise contexts, grounded in my MSc thesis at the University of Galway. This gives me an unusual ability to analyse not just what an AI system does, but whether it should do it, and how to document that decision for regulated environments.

📍 Based in Galway, Ireland · Stamp 1G (Full work authorisation) · Open to BA, PM, PO, and applied AI research roles

Business Analysis    →  Requirements Elicitation, BRDs, User Stories, Gap Analysis, Process Mapping
AI & GenAI           →  LLM Evaluation, Conversational AI, Prompt Engineering, AI Governance
Data & Reporting     →  SQL, Python, Power BI, KPI Definition, Dashboard Reporting
Delivery             →  Agile (Scrum / SAFe), Backlog Management, UAT, Stakeholder Workshops
Compliance           →  EU AI Act, GDPR, Regulatory Frameworks (MSc Thesis Research)

Python · PostgreSQL · Google Gemini AI · GitHub Actions · CI/CD

Built and deployed an end-to-end AI analytics agent, demonstrating hands-on Generative AI development from requirements definition through to live deployment.

• Conversational AI agent with multi-turn context management and intent-routing logic
• Stochastic simulation engine running 10,000+ iterations to stress-test decision logic against edge cases before go-live
• Full CI/CD pipeline via GitHub Actions for automated deployment and validation
• SQL reporting layer tracking platform KPIs and performance benchmarks

Python · FastAPI · PostgreSQL · Neon · Render

High-performance middleware bridging macro-strategic forecasting with micro-tactical fleet execution. Converts predictive insights into actionable logistics workflows.

• Dynamic Surge Pricing: Automatically adjusts supplier rates based on real-time fleet utilization and availability algorithms.
• Aggregation Engine: Consolidates multi-supplier capacity and prioritizes low-emission vehicle routing.
• Enterprise Architecture: Deployed across a multi-cloud environment (Render & Neon) with comprehensive CI/CD, global exception handling, and observability telemetry.

Python · FastAPI · UiPath · PostgreSQL · Docker · Prometheus

An enterprise-grade hyperautomation bridge designed to resolve the "digital disconnect" between modern AI orchestration and legacy, non-API freight carrier portals.

• Asynchronous Execution: Implemented a non-blocking Background Task model with 202 Accepted status patterns, decoupling high-latency RPA workflows from upstream demand signals.
• Idempotent Transaction Logic: Engineered a repository layer to prevent redundant carrier bookings, ensuring financial consistency across distributed system retries.
• Robotic Orchestration: Directly integrates with UiPath Orchestrator Queues to trigger unattended robots, translating JSON payloads into simulated UI keystrokes for automated fulfilment.
• Distributed Observability: Established full-stack tracing using X-Correlation-ID and Prometheus metrics to monitor system health and robotic success rates in real-time.

Vehicle Counting and Classification Using Convolution Neural Network International Research Journal of Modernization in Engineering Technology and Science (IRJMETS), Jul 2022 - Methodology: Led the development of a detection pipeline using YOLOv3 trained on a custom dataset of 8,342 images.

• Contribution: Integrated ORB Feature Extraction for real-time trajectory analysis, achieving >80% classification accuracy.
• Relevance: Demonstrates the application of Computer Vision to "Empirically Grounding Analytics" (EGA).