Mohammad Kamel
R & d Computer Vision Engineer @Alascom
Signup · Get unlimited contacts
WORK HISTORY
R & d Computer Vision Engineer @Alascom
Milan, IT
Implemented a vision system responsible for car seat cover quality control using Anomalib and YOLO, achieving 96% accuracy (Python, PyTorch).* Built a RAG-augmented voice command palletizing system using Whisper and LLaMA 3 to convert multilingual commands into robotic tasks using factory SOPs and task rules that eliminated operator station visits (90%) and cut the operator training time by 50%(Python).* Designed a quality assessment system responsible for measuring the diameter, height, and surface curvature of cheese with a 3D profile sensor with a measurement error less than 4mm (Open3d, Python).* Created a cloud-based visual inspection system for automated defect detection in T-nuts using Amazon Lookout, achieving 96% accuracy (AWS, Python, S3, Lambda).* Developed a 6-Dof pose estimation system for screw pick-and-place task using structured light scanner and Halcon.* Proposed a vision-guided robotic pick-and-place system using ArUco markers and 6-DoF pose estimation, achieved <5mm localization precision for shelf-mounted boxes, overcoming low-resolution Camera limitations to enable fully autonomous warehouse operations (Python, OpenCV).* Designed a conditional GAN to synthetically generate leather defects, eliminating 3+ months of on-site data collection and accelerating inspection system development.* Established a pick-and-place vision system using MobileSAM segmentation, achieving an average centroid error of ±3 mm without relying on box size constraints or model retraining.* Built a quality inspection system to detect defects on printed pasta boxes that achieved a 95% accuracy rate in identifying defective printings (Halcon, C++).* Automated gearmotor assembly monitoring using a compound-scaled object detection model, saving 300+ labor hours/month (Python, YOLOv7).
EDUCATION
Sapienza Università di Roma
Master's degree
Ferdowsi University of Mashhad
Bachelor of Science - BS
ABOUT MOHAMMAD KAMEL
Motivated Computer Vision Engineer adept at harnessing the power of Deep Learning and Machine Learning to solve complex challenges. Currently serving as an R&D Computer Vision Engineer at Alascom s.r.l, I specialize in leveraging advanced techniques and technologies to develop innovative solutions across diverse applications. My expertise spans a wide range of computer vision software libraries, including OpenCV and MVTec HALCON, with strong proficiency in Python and C++. I have hands-on experience deploying vision systems both on-premises and in the cloud, utilizing AWS services to build cloud-integrated workflows. I am also exploring Large Language Models (LLMs) to enhance automation and create intelligent, human-like interaction layers that complement vision-based systems. In my role, I engineer systems that empower robots and automated machinery to interpret visual data. This includes building high-precision solutions for quality inspection and robot guidance in industries such as food, textile, packaging, and manufacturing. These projects highlight both my technical proficiency and my drive to push the boundaries of industrial innovation. Collaboration is central to my approach. I thrive in dynamic, cross-functional teams where diverse perspectives are valued. I have supervised other vision engineers and worked closely with robotics, PLC, and software teams to deliver integrated solutions. Through open communication and shared goals, I help ensure that projects run smoothly and exceed expectations. Passionate about driving progress in both computer vision and project execution, I am committed to staying at the forefront of this fast-evolving field. With a proactive mindset and a strong focus on emerging trends, I continually seek to expand my skill set and contribute to solutions that create real-world impact.
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.