Philosophy: Build. Deploy. Measure. I specialize in creating AI that survives the "wild." My work focuses on building mobile systems for low-resource environments, collecting original datasets where none exist, and using Generative AI to turn telematics into meaningful, culturally appropriate feedback. I believe in AI that doesn't just compute, but communicates—prioritizing safety by delivering insights after the journey, not during the drive.
Currently focused on transitioning a validated smartphone-based telematics prototype into an institutionally adopted road safety tool for West Africa, featuring AI-driven driving feedback, alcohol detection, and hijack detection for fleet operations.
Thompson, I. P., Yi, D., & Reiter, E. · 2026 · 12th International Conference on Computer and Communications (ICCC), IEEE
Thompson, I. P. · 2025 · University of Aberdeen
Thompson, I. P., Yi, D., & Reiter, E. · 2025 · 30th International Conference on Automation and Computing (ICAC), Loughborough, UK
Thompson, I. P., Yi, D., & Reiter, E. · 2024 · Software Impacts
Thompson, I. P., Willie, E., Yi, D., & Reiter, E. · 2024 · AFSIC-NDU2024, Amassoma, Nigeria
Transitioning a validated smartphone-based telematics prototype into a widely adopted safety tool for fleet operations in West and Central Africa.
An open-source Android app for collecting real-time driving and sensor data in Nigerian conditions.
An end-to-end NLG pipeline that generates culturally sensitive post-trip driving feedback for Nigerian drivers.
A decision-tree classifier detecting alcohol-influenced driving from smartphone sensor data collected in Nigeria.