Aniello Francesco Prisco

Built an end-to-end EV telemetry pipeline for a luxury OEM connected car platform and developed production Bayesian models for EV battery RUL estimation from sparse, irregular signals. Implemented a two-level alert system (WARNING/CRITICAL) and robust feature engineering for on-board telemetry with missing data handling.

Designed and deployed LLM-powered document classification pipelines, reducing manual review time by 60% and delivering ~£500K annual savings. Built scalable Azure NLP pipelines with 99.9% uptime and introduced CI/CD and automated testing across the ML codebase.

Developed HPC evaluation pipelines using Dask for large-scale model benchmarking. Optimized distributed workloads on compute clusters, reducing benchmarking runtime by 50%, and contributed infrastructure libraries to improve reproducibility and scalability.

Built predictive maintenance systems that reduced equipment downtime by 40% (~£1.8M annual savings) and delivered analytics dashboards and predictive pipelines across 20+ global assets. Led an Agile team of 5 engineers, mentoring on TDD, CI/CD, and modular design.

Built backend REST APIs and microservices in Java integrated with SQL databases during a graduate program. Completed intensive training in Agile software engineering, testing, and CI/CD practices.

MSc in Quantitative Finance with specialisation in stochastic processes, Monte Carlo simulation, and numerical optimisation techniques.

BSc in Mathematics, Statistics and Probability, including a thesis on predictive modelling techniques using Bayesian inference and statistical learning.