Dmitriy Bolotov
Let's Connect
Navigation
Hi!
I’m a senior data scientist focused on applied machine learning and practical AI.
Over the past decade, I’ve worked across adtech, manufacturing, and industrial analytics, helping teams move from noisy data to working models. My experience spans predictive modeling, anomaly detection, deep learning, and LLM-based tools.
This site shares selected projects and tools I've developed or maintained.
R Packages
I worked on several open-source R packages to support machine learning model deployment and experimentation:
- pmml and pmmlTransformations – Export trained predictive models and preprocessing steps to PMML, an open standard for model deployment.
- neighbr – K-nearest neighbors for classification, regression, and clustering. [Intro]
- amelie – Anomaly detection using binary classification and maximum likelihood estimation. [Intro]
Data Science Projects
- Time Series Smoothing – Interactive comparison of six smoothing techniques for time series data, using Streamlit and Python.
- Motion Recognition – Classifying human motion from accelerometer and gyroscope data using random forest models.
- Severe Weather – Identifying the most economically and medically impactful weather events in the U.S. using NOAA data in R.