Dagim Dessalegn Haile

Developed a deep learning pipeline for detecting pedestrian crosswalks using satellite imagery. Trained YOLOv8 models on 5,138 images, achieving 87.4% mAP50–95, and applied clustering and HOG-based methods for crosswalk quality assessment.

Conducted baseline assessment of urban speed zones using traffic, land use, and socioeconomic data. Performed GIS analysis and validation of speed limits using QGIS and Google Maps, and evaluated policy frameworks for speed management.

Managed project planning, scheduling, and budgeting for road construction works. Reviewed road design plans, conducted quality control, and performed road safety audits and safety impact assessments.

Evaluated and selected construction materials for road projects. Conducted laboratory and field tests on soil, aggregate, concrete, and asphalt, and monitored material quality to ensure compliance with standards.

Performed material testing on a highway project and interpreted laboratory results for quality control. Supported contract administration and site documentation, and assisted road design using AutoCAD Civil 3D and Eagle Point.

Pursuing an MSc in Artificial Intelligence at KU Leuven, Belgium, from September 2024 to August 2026.

Completed an MSc in Transportation Science with a focus on Traffic Safety at Hasselt University, Belgium, from 2020 to 2023.

Earned a BSc in Civil Engineering from Jimma University, Ethiopia, from 2011 to 2016.