Edward S.

Edward S.

RzeszowPoland

AI & Machine Learning | Amazon Web Services, Artificial Intelligence

SUMMARY Innovative Artificial Intelligence Engineer with 7֡ years of experience in application design, development, testing, and deployment. Highly experienced in writing codes and algorithms as well as building complex neural networks through various programming languages ՄC#, C֡֡, JavaScript, Python) . Have done 10֡ big AI projects successfully and implemented 30֡ new important algorithms., PROJECTS Autonomous Driving: Trajectory Prediction and Tracking - Leveraging multi-modal data ՄLIDAR/image) to build contextual multi-agent trajectory prediction retaining scene context. - Seek to improve prediction by using agents' past movements, social interactions, scene context, and stochastic of humans. Deep Learning on Edge Devices - Deployed compressed-neural architectures ՄMobile Net v2/Shuffle Net v2Յ on edge devices to reduce computational load. - Improved inference time for face recognition at edge device while maintaining comparable
Edward S.

Edward S.

RzeszowPoland

AI & Machine Learning | Amazon Web Services, Artificial Intelligence

Specializes in
SUMMARY Innovative Artificial Intelligence Engineer with 7֡ years of experience in application design, development, testing, and deployment. Highly experienced in writing codes and algorithms as well as building complex neural networks through various programming languages ՄC#, C֡֡, JavaScript, Python) . Have done 10֡ big AI projects successfully and implemented 30֡ new important algorithms., PROJECTS Autonomous Driving: Trajectory Prediction and Tracking - Leveraging multi-modal data ՄLIDAR/image) to build contextual multi-agent trajectory prediction retaining scene context. - Seek to improve prediction by using agents' past movements, social interactions, scene context, and stochastic of humans. Deep Learning on Edge Devices - Deployed compressed-neural architectures ՄMobile Net v2/Shuffle Net v2Յ on edge devices to reduce computational load. - Improved inference time for face recognition at edge device while maintaining comparable
More than 30 hrs/week