Ahmed Bahnasy
MLOps ML Infra - Interior Sensing at Aptiv
Wuppertal, North Rhine-Westphalia, Germany
Summary
👤
Senior🎓
Top SchoolAhmed Bahnasy is an MLOps and ML infrastructure engineer focused on interior sensing at Aptiv, combining nine years of hands-on experience across computer vision, control systems, and data science. Trained as a mechatronics engineer with graduate studies at TUM, he has worked on applied research projects—from 3D object detection with GNNs to a master thesis at Helmholtz Munich—and shipped production-focused software at automotive suppliers like Valeo. His open-source contributions include implementing an AMASSLoader in the popular BlenderProc project to synthesize realistic human poses using SMPL-H, reflecting a practical blend of ML, graphics, and simulation. Equally comfortable with control‑level engineering and ML infrastructure, he brings cross-disciplinary fluency in sensors, motion models, and deployment pipelines. Outside work he pursues mountaineering and photography—having solo-climbed Germany’s Zugspitze—which underscores a disciplined, adventurous mindset that complements his technical drive.
9 years of coding experience
4 years of employment as a software developer
Student, Secondary Education, Student, Secondary Education at El-Salam Private Language SchoolMaster of Science - MS, Informatics, Master of Science - MS, Informatics at Technical University MunichStudent, Primary & Preparatory Education, Student, Primary & Preparatory Education at El-Nasr Experimental Language SchoolStudent, Engineering, Student, Engineering at The German University in CairoSemester Abroad, Semester Abroad at The German University in Cairo - Berlin CampusBachelor Thesis, Control Engineering, Bachelor Project, Bachelor Thesis, Control Engineering, Bachelor Project at Hochschule für Technik und Wirtschaft Berlin
English, German
Stackoverflow
11reputation
191reached
0answers
2questions
Github Skills (11)
blender10
3d-graphics10
python10
3d-reconstruction9
computer-graphics9
machine-learning9
pytorch9
compilation6
universe6
preprocessor6
java6
Programming languages (4)
Github contributions (5)
DLR-RM/BlenderProc
Oct 2020 - Oct 2020
A procedural Blender pipeline for photorealistic training image generation
Role in this project:
ML Engineer
ML EngineerContributions:12 reviews, 13 commits, 1 PR in 13 days
Contributions summary:Ahmed implemented a new object loader (AMASSLoader) for the BlenderProc project, enabling the loading of .obj files from the AMASS dataset. This loader generates human poses from motion capture data by integrating the SMPL-H body model. The user's contributions involved loading data from the AMASS dataset, extracting pose and shape parameters, and writing the generated poses to .obj files. The user also contributed to the integration of the SMPL model within the BlenderProc environment, utilizing the human_body_prior library.
structure-from-motionpythonblenderfeature-matching3d-engines
abahnasy/leetcode
Jun 2021 - Dec 2021
Contributions:366 pushes in 5 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.