maaz
Research Scientist
Beijing, China
Summary
👤
SeniorMuhammad Maaz is a research scientist and final-year PhD candidate specializing in multimodal large language models and long-form video understanding, based at MBZUAI with advising from Salman Khan and Fahad Khan. He has eight years of experience bridging applied computer vision and research, with production deployments on edge devices and industry internships at Meta where he developed open, competitive MLLMs like PerceptionLM/PLM. His work includes Video-ChatGPT and a 100K high-quality video-instruction dataset plus the first quantitative evaluation framework for video conversation models, highlighting a rare combination of dataset engineering, evaluation rigor, and model innovation. Comfortable moving models from lab to product, he has built CPU-optimized vision systems for real-world retail and traffic applications and mentors students as a teaching assistant. Based in Beijing, he brings deep academic training (MS with 4.0 GPA) and hands-on engineering that enables reproducible, open research that competes with closed-state-of-the-art systems.
8 years of coding experience
Github Skills (15)
robot7
gps6
robotics6
ros6
visual-odometry6
arduino5
localization5
opencv4
raspberry-pi4
line-segment4
slam4
deep-learning2
algorithm2
convolution2
graph1
Programming languages (4)
Github contributions (5)
maazmb/LEP-Hybrid-Visual-Odometry
Dec 2019 - Mar 2021
Contributions:3 commits, 1 push, 1 branch in 1 year 3 months
maazmb/rotation_extraction
Mar 2018 - Dec 2019
Contributions:2 pushes, 1 branch in 1 year 9 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.