I'm an R&D Engineer in Machine Learning. I can help you from debugging/reviewing your code to implementing code of an academic publication and serving it on production. I also have experience in AI/Data Science consulting for startups and SMEs.
My experience varies across prototyping ML models from academic models and designing new algorithms, to designing ML systems in production. As for the tools, I work with python data science stack (numpy/pandas/matplotlib/scikit-learn) and the deep learning frameworks (keras/tensorflow/pytorch). My main focus for the past few years has been computer vision and audio applications using deep learning.
Video Surveillance R&D: End-to-end ownership in classification problems applied to video surveillance using a combination of deep ...
Video Surveillance R&D: End-to-end ownership in classification problems applied to video surveillance using a combination of deep learning and traditional computer vision. Implementing code for relevant Deep Learning archi- tectures from academic publications. Training, improving, and debugging models (eg., CNNs, Vision Transformers). Data set management: data collection, using available deep learning models (eg., object detection, segmentation, and classification models) to clean/enrich the existing data. Model compression (eg. Knowledge Distillation). Designing novel algorithms and prototyping - when successful: integrating to the product SDK.