Home

Hey! I'm Maged - researcher, founder, and professor working at the intersection of AI, computer vision, and transportation.

I'm the Founder & CEO of ApexAI, where we're building AI-powered navigation for visually impaired people. Our app runs four transformer models entirely on your phone. Unlike existing solutions that cost $3,500–$4,500 (OrCam, Envision), rely on cloud processing with 1–5 second latency (Seeing AI), or stream video to strangers (Be My Eyes), ApexAI runs entirely on-device at <200ms latency with zero cloud dependency. It's live on iOS and we have a U.S. patent pending.

My research has led to 20+ publications in venues like IEEE CVPR, WACV, and ASCE journals, with over 1,500 citations on Google Scholar. I've been fortunate to deliver keynotes at ICCV 2025, WACV 2025, and Oak Ridge National Laboratory, and to secure over $1.4M in research funding from the U.S. DOT, NSF, and Meta.

When I'm not building products or chasing deadlines, I'm usually running or skiing!

Selected Work

Apex Vision App (Live on iOS) - Three modes: Finder for object localization, Safety for walkable-surface overlay with hazard alerts, and Describe for dense scene captioning. The only product offering real-time pedestrian navigation. U.S. Patent Pending.

Presenting “Are We Ready to Deploy Agents on the Grid or Plant?” at the Artificial Intelligence for Robust Engineering and Science (AIRES-6) Workshop hosted by Oak Ridge National Laboratory - Sep 2025.

Delivering the keynote “Research, Develop, Deploy” at the International Conference on Computer Vision (ICCV) in Honolulu, HI. - Oct 2025.

Cooperative Driving Automation (CDA) Platform: Developing a 1/10th-scale high-fidelity autonomous vehicle platform for cooperative driving research. The platform bridges simulation-to-reality gaps for V2X communication, cooperative perception, and platooning experiments.

Presenting at IEEE CVPR 2024 on Vision-Language Models for Dense Video Captioning of traffic safety events. 6th place, NVIDIA AI City Challenge Track 2: VLM for Dense Captioning - June 2024.

Opening keynote on “Innovations in Challenging the Unexpected in Dashcam Videos” at the IEEE Winter Conference on Applications of Computer Vision, Tucson, AZ. - Mar 2025.

Research on COVID-19’s impact on St. Louis bus services featured on St. Louis National Public Radio - research Dec 2020.

Recognized as Outstanding PhD Student by the College of Engineering, University of Missouri. One of five annual awardees across all engineering departments: Outstanding Ph.D. Student Award - Oct 2021.

Designed a Graph Convolution–Gated Recurrent Unit (GC-GRU) architecture for traffic forecasting using connected vehicle data. Achieved a 50× speed-up in big-data ETL pipelines through GPU-accelerated processing with RAPIDS APIs. Built an interactive, speech-driven web application (NOCONS) for transportation planners to visualize and predict traffic congestion: NOCONS - Nov 2022.

Presenting 1st place research award on connected vehicle data for real-time congestion detection at the ITS Heartland Annual Meeting- Nov 2021.

My research on Connected Vehicles impact on congestion detection winning first place in ITS Heartland - Jan 2022.

Delivering my 3-Minute Thesis presentation about Generative AI at MU's Research Symposium - Oct 2022.

Year-end celebration as ASCE Student Chapter President, American University of Sharjah. May 2015.

Welcoming ASCE Global President Dr. Robert Stevens (right) to AUS as the first Student Chapter President to host a sitting ASCE president. On the left is Dr. Osman Akan, AUS - CEE Department Head. - May 2015.

I had the honor of serving as the ASCE Student Chapter President for the 2014-2015 academic year at AUS.

High School project: Automated Car Wash System - June 2010.

4th place in the World Red Bull Can Car Racing Competition - Dec 2011.

Selected Work

Get in touch at magedshoman@gmail.com