Dmitry Migunov
Data Scientist @YIT - Urban developer and construction company
Signup · Get unlimited contacts
WORK HISTORY
Data Scientist @YIT - Urban developer and construction company
Helsinki, FI
Developed a system that uses facial recognition technology to identify individuals and prevent unauthorized access to construction sites. As a result, the system reduced unauthorized site entries by over 70%, significantly improving overall security and compliance.The system consists of several components:1. A head rotation analysis module built using OpenCV2. A facial comparison module based on deep learning techniques suchas DeepFace and Faiss, which allows for fast face searches in a databaseTechnologies: python, AWS, SQL, opencv, deepface, faiss, pm2, docker• Developed a system that can recognize text data from photos and scans of documents, which has helpedreduce the workload of specialized professionals by 73.5%.The system consists of several components:1. A module that analyzes the rotation of the document image2. A module to determine the document page3. A segmenter that divides the page into blocks4. A text recognition module that recognizes text in the segmented parts of the passport using OCR (OpticalCharacter Recognition)Technologies: python, PyTorch, YOLO 11, SQL, OpenCV, EasyOcr, Transformers, labelstudio• Developed a RAG solution for the LLM based on the company’s internal documents. This solution hashelped reduce support requests by 30.3% initially and then by an additional 61.1%.Technologies: python, DeepSeek, ollama, Qdrant, BERT, docker, openWebUi, SQL
EDUCATION
ITMO University
Master
ITMO University
Information Science and Computation Technology
ABOUT DMITRY MIGUNOV
A data scientist with four years of experience in commercial development across data science, deep learning, computer vision, and LLMs. I have successfully implemented end-to-end machine learning solutions, covering requirement gathering, data preprocessing, model development, deployment, and monitoring in production.Worked as part of collaborative teams of 2 to 8 people, both remotely and in office settings, where I closely interacted with engineers, analysts, and product teams. This allowed me to gain valuable experience in team-oriented development, code reviews, and agile workflows, contributing to faster and more robust ML solution delivery.Skills: Computer Vision, NLP, LLM, AWS, Machine Learning, Python
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.