Hi, I am Gleb

I am Team Lead of ML Matching Team
I hold a master's degree in computer science in ITMO University
I love statistics
I built AI project called Brandy
I love coming up with new ideas

Gleb Drozdov

Machine Learning Team Lead at Ozon

I am a passionate data scientist who wants to make our world a better place using Artificial Intelligence. I`ve worked 5 years as Data Scientist and implemented more than 10 AI projects.

Machine Learning

MLOps

Python

Java

Skills

Machine Learning

Main field of my professional development.

Natural Language Processing

Most experienced in this subfield of machine learning. Implement topic modeling, sentence clusterization, Q&A systems, Intent Classification, Sentence Generation models, e t.c.

Python Development

Using as the main language for professional development. Capable of writing scalable, testable, and maintainable program. Write a lot of scraping scripts for dataset collections

MLOps and Data Engineering

Have a big experience with tools such as Spark, Hadoop, Airflow, Kafka, Grafana, MLFlow

Java

Implement backend and mobile apps. Use it to write and maintain high-load systems.

PyTorch

Implement and fine-tune a lot of Deep Learning models, such as Generative Adversial Networks, transformers, BERT, GPT-2.

Linux

Using as the main operating system. Capable of writing bash/shell scripts. Work with ML models on remote machines

Experiences

Machine Learning Team Lead

Ozon

August 2022 - Present, Moscow

Ozon is a leading market player and one of the most valuable Russian internet companies, according to Forbes.

Responsibilities:

Develop microservice architecture for real-time matching
Implement algorithms for finding products which are not represented on the marketplace
Optimize production pipelines
Leading a team of 6 ML engineers
Communications with customers

Machine Learning Engineer

Ozon

September 2021 - August 2022, Moscow

Ozon is a leading market player and one of the most valuable Russian internet companies, according to Forbes.

Responsibilities:

Develop ML Matching algorithms, which find the same products from a competitor marketplace.
Implement several ML model for products matching with precision 95% and recall 25%
Train and integrate in pipeline several NLP models (BERT, fasttext, BM25)
Implement a lot of algorithms analytics in Grafana and Slack

Data Scientist

SberBank

April 2021 - September 2021, Saint-Petersburg

SberBank is the largest bank in Russia, Central and Eastern Europe, and one of the leading financial institutions worldwide.

Responsibilities:

Implement Recommendation System based on client-manager dialogs

Data Scientist

iPavlov is the global project for conversational artificial intelligence research and development, performed by the Neural Networks and Deep Learning Lab of Moscow Institute of Physics and Technology (MIPT) and supported by the National Technological Initiative (NTI) and PJSC Sberbank.