EMMANUEL EZENWERE

– Architected and owned end-to-end delivery of a production ML recommendation system serving 90 million global customers, driving £55M net annual revenue contribution, owning the full lifecycle from data engineering, feature design, and model development through to MLOps deployment and monitoring. – Designed and deployed bot detection and fraud mitigation models, protecting platform integrity across high-volume, adversarial environments with continuous model drift monitoring. – Implemented responsible AI infrastructure using the Aequitas fairness framework, achieving measurable reductions in model bias across protected demographic groups, alongside SHAP-based explainability enabling regulatory-grade model transparency. – Designed and implemented an LLM-based AI agent for Source of Funds (SoF) verification, architecting multi-step reasoning and tool orchestration to automate compliance workflows, with robust validation, guardrails, and evaluation strategies for reliable decision-making. – Supported technical hiring by defining evaluation criteria and participating in interviews to recruit AI talent and build out the organisation's ML capabilities.

– Led secure migration of terabytes of confidential clinical research data from Azure to GCP, designing cryptographic validation pipelines to ensure zero data loss and audit-grade integrity. – Architected HPC environments optimised for low-latency ML workloads, materially reducing migration timelines and unblocking research team velocity. – Established scalable, configurable cloud infrastructure that became the standard workspace for downstream ML research workflows.

– Designed and deployed mission-critical AI software for 5G Open RAN network optimisation, directly improving network efficiency and reducing operational costs at national infrastructure scale. – Architected a real-time data pipeline processing 2,000+ datapoints/second using Apache Kafka, Apache Spark, Databricks, and distributed SQL, from XML extraction through stream processing to persistent storage. – Built and deployed ML-powered xApps on the Near Real-Time RIC, implementing automated network recalibration, dynamic resource allocation, energy optimisation, and load balancing. – Developed cloud-native rApps with real-time Power BI dashboards, predictive maintenance models, automated policy generation, and QoS management. – Developed an NHS diabetes prevalence prediction system across UK LSOAs: designed ETL pipeline for NHS patient data, built validation workflows, and deployed ML models achieving 85% prediction accuracy. – Established enterprise-grade ML engineering practices across the programme, leading cross-functional teams and managing key stakeholder relationships at government level.

– Founded and led the technical organisation building autonomous drone logistics for healthcare in Sub-Saharan Africa. – Led a multidisciplinary engineering team to design, build and deploy autonomous cargo drones improving medical supply logistics 5x in rural Nigerian healthcare centres, architecting automated flight planning and control systems with AI Planning Algorithms for reliable edge-based autonomous operations. – Developed edge computing solutions for real-time people detection, facial recognition, and demographic analytics at 95% accuracy / 30 FPS on Nvidia Jetson hardware with GPU acceleration. – Won the Total Young Entrepreneur Prize; recognised by Oxford Business Group as a leading voice in drone logistics disruption.

– Retrained a CNN (TensorFlow Inception V3) via transfer learning to classify face shapes at 95%+ accuracy on a custom dataset. – Built a full computer vision pipeline using Dlib and OpenCV for facial landmark extraction, image transformation, and hairstyle fitting. – Deployed as a web service with REST APIs, handling 1.5M+ API requests in production.