(remote) - Developed ML models of clustering and assortment matching for e-commerce (NLP, BERT, Glove). - Developed a multivariate analysis algorithm for price segmentation and potential assessments, allowing businesses to run new product lines faster and more clearly define the positioning in the market and get more revenue. - Established a data governance system for the product marketing team, from design to implementation and active utilization. - Implemented an assortment coverage gap assessment system. - Developed demaind forecast prototype based on Prophet. - Through independent interaction with stakeholders and business customers, I initiated and continue to promote a cross-functional project.

- Developed a navigation chatbot for a high-traffic exhibitions (2 mil. visitors) website and mobile app. Сontributed to System Design the app base on API microservices. - Implemented an entities extraction service and designed and optimized the search engine, developed search logging, data preparation service (data annotation and summarization). - Notably, crafted an intent detection service using Large Language Models, enhancing the chatbot's performance and user experience.

(remote USA) - Developed a therapist recommendation module using Flask and a semantic search approach, including data preparation and few-shot classification for intent detection (TaiChi/USLP). - Developed chat-bot prorotype based on GPT-3 and few-shot learning, BERT-like model as well as implemented context memory function, GPT-3 API integration, and OpenAI Best Security Practices for the app. - Managed the project and collaborated with member support and engineering teams to meet all project deadlines.

Prediction of voltage consumption. I prepared raw data, made EDA, implemented a SARIMA model and an LSTM predictive model for time series extracted from logs and got high quality prediction (few articles implementation).

(project work) Implemented Anomaly Detection models for IoT devices: - one based on an Autoencoder and - one based on LSTM autoencoder NN. Developed the dashboard for visualization the results used streamlite library.

startup (remote England, Portugal, Russia) We tried to analyze the text for the psychological state of the author, looking for anomalies in the text. And developed logs anomaly detection service/plugin for this. - Found train datasets (ENRON), prepared, cleaned, analyzed, developed and check hypotheses, developed the pipeline, and visualized the results and created dashboard (streamlit lib). - Identified 26 lexical, semantic and other text features. - Several features I developed have been added to our own NLP library for the product.

-Made EDA of AzureAdLog Dataset and AzureEmail Dataset. -Develop implementation of the anomaly detection algorithms for these logs. -Develop Log Anomaly Detection web-app used streamlit library. -Research many information resources and datasets for training ML models. -Manual labeled moreover 500 news for supervised machine learning model. -Found over 60 correlations between different retail factors (search and analysis datasets with close features).

- Developed a comprehensive Anti-FROAD program that stopped a wave of viral activity in the office. - My FROAD security increase approach has been extended to other regional offices of the PFR (Proposed an approach to organizing an access management system (Oracle. OIMS)) - Developed the role model with stakeholders, which eliminated the possibility of exceeding authority (FROAD method) And access time has been reduced from 2 days to several minutes - Deployed the Device Control System (based on Kaspersky Endpoint Security), which ensured the fulfillment of several legislative requirements at once. - Administer VPN network, control firewalls, anti-virus system and IAM settings, solve network issues, work with MS SQL. Handled with froad events.

http://www.spiiras.nw.ru/, PhD in Information Tethnologies

Information systems and technologies, Cybersecurity specialist, Master degree

Coursera, TensorFlow on Google Cloud

Coursera, Google Cloud Big Data and Machine Learning Fundamentals

LinkedIn , Intermediate SQL for Data Scientists

LinkedIn, Git Essential Training: The Basics

Coursera, Data Analysis

Coursera, Finding a structure in the Data

Coursera, Mathematics and Python for Data Analysis

Coursera, Supervised Learning for Data Science

Sberbank, finteth

IBM, has attended the above course and has attained the requisite standard of perfomance