I’m Mariya Sytsko, a Machine Learning Engineer with hands-on experience in computer vision, OCR systems, and intelligent document processing. My journey in AI started with data annotation and dataset optimization, where I built a strong foundation in understanding how high-quality data directly impacts model performance. Over time, I transitioned into full-scale machine learning engineering, focusing on building practical, scalable solutions for real-world industrial challenges. Previously, I worked as an ML Engineer at CAE Technology, where I developed and optimized computer vision pipelines for visual similarity analysis, anomaly detection, segmentation, and OCR-based information extraction. My work combines both deep learning and classical computer vision techniques, allowing me to build systems that are not only accurate but also efficient and robust. I’ve also designed LLM-powered document intelligence systems using GPT-based extraction, prompt engineering, and RAG pipelines to transform unstructured data into structured insights. Previously, I worked as a Data Analyst at CAE Technology, where I specialized in dataset annotation, validation, and preprocessing for computer vision projects. This experience gave me a deep understanding of the entire ML lifecycle — from raw data preparation to model optimization. What makes me stand out is my ability to combine research thinking with practical engineering. I’m comfortable moving between experimentation, algorithm design, and deployment. Whether it’s improving pipeline speed by 2.2×, replacing fragile model-dependent logic with geometry-based solutions, or benchmarking multiple approaches to find the best fit, I focus on building solutions that are both technically strong and business-oriented. My core expertise includes Python, FastAPI, OpenCV, PyTorch, TensorFlow, OCR technologies, LLMs, and RAG systems, with a strong passion for solving complex problems in computer vision and machine learning.…I’m Mariya Sytsko, a Machine Learning Engineer with hands-on experience in computer vision, OCR systems, and intelligent document processing. My journey in AI started with data annotation and dataset optimization, where I built a strong foundation in understanding how high-quality data directly impacts model performance. Over time, I transitioned into full-scale machine learning engineering, focusing on building practical, scalable solutions for real-world industrial challenges. Previously, I worked as an ML Engineer at CAE Technology, where I developed and optimized computer vision pipelines for visual similarity analysis, anomaly detection, segmentation, and OCR-based information extraction. My work combines both deep learning and classical computer vision techniques, allowing me to build systems that are not only accurate but also efficient and robust. I’ve also designed LLM-powered document intelligence systems using GPT-based extraction, prompt engineering, and RAG pipelines to transform unstructured data into structured insights. Previously, I worked as a Data Analyst at CAE Technology, where I specialized in dataset annotation, validation, and preprocessing for computer vision projects. This experience gave me a deep understanding of the entire ML lifecycle — from raw data preparation to model optimization. What makes me stand out is my ability to combine research thinking with practical engineering. I’m comfortable moving between experimentation, algorithm design, and deployment. Whether it’s improving pipeline speed by 2.2×, replacing fragile model-dependent logic with geometry-based solutions, or benchmarking multiple approaches to find the best fit, I focus on building solutions that are both technically strong and business-oriented. My core expertise includes Python, FastAPI, OpenCV, PyTorch, TensorFlow, OCR technologies, LLMs, and RAG systems, with a strong passion for solving complex problems in computer vision and machine learning.WWWW…