My professional story began with an investigation of non-volatile transistor memory cells in electronics. Later, recognizing its potential social impact, I shifted my focus to biomedical engineering. There, I devoted about six years to researching mechanical circulatory support technology designed for patients with end-stage heart failure. During that time, I developed a cardiovascular system model to simulate interactions with circulatory support in different heart failure conditions and came up with a method to build blood pump models for the physiological control of circulatory support. This research resulted in the publication of 17 scientific papers, contributed to securing three long-term applied research programs, and culminated in the defense of a PhD thesis. My background enhanced my skills in mathematical modelling and optimization using Python programming language, and gave me hands-on experience with complex long-term projects, laying a foundation for success in my data science role.Working in data science has unintentionally improved my soft skills, as collaboration with cross-functional teams requires considerable ability and effort to communicate results effectively, ensure proper implementation and use of machine learning solutions, and complete projects despite a series of failures. It has certainly enhanced my analytical, programming, and machine learning skills. These range from daily SQL use - including collection, preliminary analysis, and feature engineering - to employing feature selection techniques and gradient boosting, all complemented by efficient Python code.Currently, I apply these skills to combat fraud at T-Bank, protecting clients' finances and reducing the company’s losses. Over the course of four years, I have gained a deep understanding of the trade-offs inherent in machine learning solutions in this domain, in addition to working with various fraud schemes and gaining financial data literacy. I find this work fascinating and I am eager to grow my career in this field. I am currently open to job offers from international companies focused on financial fraud prevention using modern machine learning techniques on a large scale. I am ready to relocate, and visa sponsorship is required.If you’ve read this far, I suggest you connect with me whether you are interested in fraud prevention or in data science in general. After all, two heads are better than one, and it's more fun together. ----- Medium: https://medium.com/@dspetukhov- Github: https://github.com/dspetukhov- Telegram: @dspetukhov
-
Data ScientistTinkoffYerevan, Am -
Data ScientistTinkoff Jan 2020 - PresentMoscow, Russia- Developed and helped to deploy an antifraud model for card-to-card transactions in mobile bank.- Developed, deployed, and continuously improved a model to identify clients who abuse loyalty programs.- Developed the first MVP of a microservice for spam prevention in the company’s sign-on system with a peak load of approximately 300 rps using FastAPI, k8s and PostgreSQL. Developed models for the service and deployed them using Redis.- Developed several program components for internal MLOps tools, including one that optimized the feature selection process by enabling preliminary feature selection based on various criteria and eliminating the overhead of unnecessary feature storage. Contributed to optimizing and refining other components.- Optimized production scripts for the antifraud models, reducing process execution time from approximately 220 minutes to 60 - 80 minutes, depending on database load. Additionally, memory consumption decreased from around 30 GB to 100 MB.- Facing the inability to model training on large datasets, implemented a solution for model training with gradient boosting libraries for Spark in a Hadoop cluster, demonstrating an innovative approach to big data processing.- Contributed two articles on machine learning for the corporate Medium blog and conducted 40 interviews in the ML base section. - Facilitated weekly online meetings with the team responsible for the single sign-on system, ensuring close collaboration throughout the development of the spam prevention service. This initiative in some measure helped to launch this project within fourteen months, overcoming several significant failures.- Accomplished an in-house online negotiation course, ranking third out of fifty participants.About: Tinkoff is a Russian commercial bank, which is the second largest provider of credit cards in Russia, and is the world's largest digital bank, as measured by number of customers. -
Data AnalystWeborama May 2019 - Dec 2019Moscow, Russia- Provided various insightful analytics to customers using the company's online advertising data.- Drafted presentations and participated in meetings with customers.- Documented the main analytical products offered by the company as part of reducing technical debt.About: Weborama is a pioneer AdTech player in semantic artificial intelligence in France and Europe, helping agencies, publishers and advertisers in the digital advertising market. -
Research ScientistNational Research University Of Electronic Technology (Miet) Sep 2012 - May 2019Zelenograd, Moscow, Russia- Developed a cardiovascular system model that can simulate different heart failure states with mechanical circulatory support using Python and its scientific libraries. This work contributed to winning and defending several special-purpose applied grant programs through the writing of reports and 16 scientific papers indexed by Scopus and WoS, with one being published in a Q2 journal.- Secured a three-year scholarship from the President of the Russian Federation in 2016, being one of the 129 selected grantees out of 776 applications. Successfully completed the scholarship program in 2018.- Defended the institute's first thesis on ventricular assist devices.About: National Research University of Electronic Technology is a Russian technical university specializing in the field of microelectronics, information, and computer technologies, founded in 1965 in Zelenograd, Moscow. -
Research AssistantАо «Ниимэ» Oct 2010 - Apr 2013Zelenograd, Moscow, Russia- Developed guidelines for the technological simulation of non-volatile memory cells.- Created materials for training courses for bachelors and masters students.About: NIIME JSC is a leading center in Russia that conducts scientific and technological research, as well as development work in the field of micro- and nanoelectronics.
Dmitry Petukhov Education Details
Frequently Asked Questions about Dmitry Petukhov
What company does Dmitry Petukhov work for?
Dmitry Petukhov works for Tinkoff
What is Dmitry Petukhov's role at the current company?
Dmitry Petukhov's current role is Data scientist.
What schools did Dmitry Petukhov attend?
Dmitry Petukhov attended National Research University Of Electronic Technology (Miet), National Research University Of Electronic Technology (Miet), Voronezh State Technical University.
Not the Dmitry Petukhov you were looking for?
Free Chrome Extension
Find emails, phones & company data instantly
Aero Online
Your AI prospecting assistant
Select data to include:
0 records × $0.02 per record
Download 750 million emails and 100 million phone numbers
Access emails and phone numbers of over 750 million business users. Instantly download verified profiles using 20+ filters, including location, job title, company, function, and industry.
Start your free trial