Data Scientist
Resume:
Dmitry Khadmiev
Kiev, Ukraine
mailto: *********@*****.***
SUMMARY
Over 10 years of experience in data science and AI
Deep knowledge of classification, regression, cauterization and optimization methods
Excellent analytical ability in delivering data science projects
Highly skilled in R, SQL, Python, C#
Great ability to learn and use any programming language
Ability to solve unusual tasks and find non-standard solutions
International experience – Australia, Canada, US, UK, Brazil, Argentina
EXPERIENCE
2008, November – present Labs Zoral,
Data scientist
Fintech projects:
-Developed approach for debt collector’s quality prediction with non-linear algorithms (SVM, random forest) and boosting technologies
-Feature selection automation based on information theory methods
-Development of scorecards for financial projects (full cycle of predictive model development: data preprocessing, feature selection, tuning and testing of linear and non-linear algorithms)
-Analysis of users’ behavior on the web-form (with application of the time series methods)
-Automation of financial transaction preprocessing including analysis of text in transaction description, time series analysis for regular patterns detection, transaction classification.
-Sparse auto encoder development
-Extraction of unsupervised patterns from unstructured data
-Frequency analysis using DFT and wavelets
-Survival analysis
-Development of the distributed feature extraction system
-Automation of features stability and predictability analysis
-Pricing optimization using evolutionary methods
-Development of cross-country scorecards (transferring models from one country to another)
System safety and virtualization:
-Development of data preparing system for custom-length data vectors
-Development of system development for software behavior classification
-Development of intersections/absorption detection system for declarative structures (Challenging and non-standard NP-full task)
-ANN, Decision trees, probability methods, clusterization and boosting methods engine development using SGE (Number of different algorithms with single configuration system.)
-Cloud training and execution AI system development
-OCR system for graphical links and contact information on web pages
2007, August - 2008, November “Softdev spb”
Software developer
Development and support ERP/CRM system on ASP.Net
2004, September – 2007, August “NB-Systems”
Technical director of the brunch
-Migration of the transport monitoring system from Delphi to the Java and C#
-Midlet client development
-Development of the ANN and genetic optimization frameworks in JAVA and C++ (two separate frameworks with single configuration system)
-Development of the genetic floating-point algorithm for pricing optimization
-Development of the price prediction system for energy exchange
-Development of the non-linear neural network correlation searching system
-Development of the short-term prediction system for downstream hydropower
-Development of the sensor data filtration and anomalies detection system
-Development of the Russian language syntactic analysis system development and start of semantic analysis
-Genetic optimization of the mid-term operating modes of hydropower stations cascade
OWN PROJECTS:
-Electricity power generation stock market short-term prediction system
-Risk of generating equipment failure prediction system
-Analytic tool for feature analysis
-General prediction system. (Tool for analytics. Interesting as end-user project.)
-Delivery optimization system (based on the dynamical graphs and multi-criteria optimization)
-Multi-project integrated ML, test, QC and business logic clustered system.
SKILLS:
Algorithms: Artificial neural networks, SVM, Decision trees, Boosting methods (general method implementation and specific methods usage (XGBoost, random forest, c-forest, GLM net, etc.), evolutionary algorithms, imitation modeling, information theory, statistics, probability theory.
Programming languages: Python, .Net, R, C++, Java.
Languages: English(advanced), Russian (native fluency), Chineese (beginner)
Software: Perforce, SVN, CVS, TFS, PostegreSQL, MSSQL
Operating Systems: Windows
Big data stack: Map reduce, Hadoop, Spark, Hive
EDUCATION: 1999-2009 KSTU, Russia, Kazan
Engineer
Contact this candidate