Dmitriy K.

I'm a Computer Vision Engineer with 5 years of experience in computer vision and 7 years in data science and machine learning. I build computer vision systems and AI agents for businesses: visual inspection, object detection and tracking, video analytics, document processing, RAG assistants, and LLM agents. My parking violation platform automated 64% of manual inspection workflows, and my proposal matching agent raised the client reply rate from 2.0% to 4.4%. As a Computer Vision Engineer, I work with Python, PyTorch, TensorFlow, Keras, OpenCV, and C++. I use deep learning and machine learning for image processing and video tasks: image classification, object detection and tracking, image segmentation, and pose estimation. I have delivered computer vision projects in sports analytics, smart parking, video surveillance, drone imagery, and garment measurement. As a Computer Vision Engineer, I built a baseball video analytics pipeline with 91% player detection accuracy and 94% tracking accuracy, a smart parking platform that automated 64% of manual inspection, a garment measurement app that measures clothes from one photo, and a drone pipeline that finds tortoises in aerial imagery with 0.94 mAP50. These deep learning and image processing systems run on OpenCV and PyTorch, and I own them from data and annotation to deployment. As an AI Agent Developer, I build AI agents, AI bots, and RAG systems with LangChain, FastAPI, vector databases, and LLM APIs such as Claude and Gemini. I connect them to your documents, databases, and internal tools, so the agent retrieves the right data and takes action. I also combine computer vision with LLMs to turn images, scans, and documents into structured, validated data. As a machine learning engineer and AI engineer, I choose the model for your constraints: latency, hardware, dataset size, and business risk. I build custom neural network architectures when ready models do not fit, run model evaluation from the start of the project, and optimize inference for production. I use neural networks, natural language processing, and supervised and unsupervised learning. I also work as a data scientist and data analyst: data analysis with Python and SQL, data visualization, feature engineering, model metrics, and reports that non-technical teams can read. When the project needs an interface, I build it with React, Node.js, and JavaScript and connect it to Python APIs. You get a working product, not only a model. You get a computer vision or AI agent system that runs in production: data pipeline, trained model, API, deployment, documentation, and a support period. I work in clear stages: discovery, development and QA, deployment, and support. Send me your use case, your data, and your constraints. As a Computer Vision Engineer and AI Agent Developer, I will reply with a proposed architecture, timeline, and cost for your computer vision, AI agent, or machine learning project.