
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Machine Learning Engineer & Researcher | Kaggle Master
Welcome to my profile. My focus is application of Machine Learning to wide spectrum of problems including medicine, business, science, AI, image analysis, information retrieval and understanding. I am a M.Sc. in CS actively participating in machine learning challenges on kaggle.com, a popular platform that holds predictive modeling competitions. Kaggle Master ranked 129 of 66,000 in global Kaggle ranking. Winner of the "Physics Prize: HEP meets ML Award" from CERN. Experienced with Deep Learning, Neural Networks, Support Vector Machines, Ensembling, Gradient Boosting Trees. Favorite tools are Python, Scikit-learn, Pandas, Theano, Caffe, Keras.
National Research University of Electronic Technology (MIET)
M.Sc. in Computer Science and Microelectronic Devices
January 1, 1988 – January 1, 1994
School of Mathematics at Lomonosov Moscow State University
Mathematics
N/A – Present
School of Physics at St.Petersburg State University
Physics
N/A – Present
Neuromation
Senior Researcher
April 1, 2018 – Present
Self -Employed
Machine Learning projects
December 1, 2007 – Present
.
Uniastrum Bank (and affiliates)
Deputy Chief for operations on international markets
July 1, 2002 – October 1, 2007
Moscow, Moscow City, Russia
Pediatric Bone Age Assessment Using Deep Convolutional Neural Networks
October 1, 2017 – Present
Develop Deep Learning Bone Age Assessment (BAA) system using data from the 2017 Pediatric Bone Age Challenge. The organizer: The Radiological Society of North America (RSNA) Technologies used: Convolutional Neural Nets, Python, Keras. Publication: https://doi.org/10.1101/234120
Text Normalization Challenge - Russian Language
September 1, 2017 – November 1, 2017
Convert Russian text from written expressions into spoken forms. The sponsor: Google.
Text Normalization Challenge - English Language
September 1, 2017 – November 1, 2017
Convert English text from written expressions into spoken forms. The sponsor: Google.
NIPS 2017: Targeted Adversarial Attack
July 1, 2017 – October 1, 2017
Develop an adversarial attack that causes image classifiers to predict a specific target class. Sponsor: Google Brain Github: https://github.com/alexander-rakhlin/NIPS-2017-Adversarial-contests
NIPS 2017: Non-targeted Adversarial Attack
July 1, 2017 – October 1, 2017
Imperceptibly transform images in ways that fool classification models. Sponsor: Google Brain Github: https://github.com/alexander-rakhlin/NIPS-2017-Adversarial-contests
NIPS 2017: Defense Against Adversarial Attack
July 1, 2017 – October 1, 2017
Create an image classifier that is robust to adversarial attacks. Sponsor: Google Brain Github: https://github.com/alexander-rakhlin/NIPS-2017-Adversarial-contests
Sea Lion Population Count
March 1, 2017 – June 1, 2017
25 place of 385 teams (top 7%). Developed algorithms which accurately count the number of sea lions in aerial photographs. The sponsor: NOAA Fisheries. Python, Theano, Keras Deep Learning environment.
Rental Listing Inquiries
February 1, 2017 – April 1, 2017
73rd of 2488 teams (top 3%). Predicted the number of inquiries a new rental listing receives. The sponsors: Two Sigma and RentHop. Python, Scikit-learn, Boosting trees (XGBoost, LightGBM)
NIH Seizure Prediction
December 1, 2016 – Present
76 place of 478 teams. Predict seizures in long-term human intracranial EEG recordings. Seizure forecasting systems have the potential to help patients with epilepsy lead more normal lives. In order for electrical brain activity (EEG) based seizure forecasting systems to work effectively, computational algorithms must reliably identify periods of increased probability of seizure occurrence.
Ultrasound Nerve Segmentation
August 1, 2016 – Present
41 place of 923 competitors (top 5%). Identify nerve structures in ultrasound images of the neck. Accurately identifying nerve structures in ultrasound images is a critical step in effectively inserting a patient’s pain management catheter. In this competition, Kagglers are challenged to build a model that can identify nerve structures in a dataset of ultrasound images of the neck. Doing so would improve catheter placement and contribute to a more pain free future. Python, Theano, Keras Deep Learning environment.
Facebook V: Predicting Check Ins
July 1, 2016 – Present
38th place among 1212 teams (top 10%). Predict which place a person would like to check in to. Technologies used: Pandas, XGBoost, Scikit-learn. The sponsor: Facebook.
Yelp Restaurant Photo Classification
April 1, 2016 – Present
22 place among 355 competitors (top 10%). Predict attribute labels for restaurants using user-submitted photos. In this competition, Yelp is challenging Kagglers to build a model that automatically tags restaurants with multiple labels using a dataset of user-submitted photos. Currently, restaurant labels are manually selected by Yelp users when they submit a review. Selecting the labels is optional, leaving some restaurants un- or only partially-categorized. Tools and methods: Caffe Deep Learning Framework, Python, Scikit-learn, Pandas, H5py, Theano, Keras, XGBoost. Github page: https://github.com/alexander-rakhlin/Yelp Article: https://www.linkedin.com/pulse/article/what-restaurant-would-your-computer-like-go-alexander-rakhlin
CNN for Sentence Classification
March 1, 2016 – Present
Convolutional network for movie reviews sentiment analysis. Python, Theano, Keras
The Allen AI Science Challenge
February 1, 2016 – Present
40 position among 170 teams (top 25%) The Allen Institute for Artificial Intelligence (AI2) is working to improve humanity through fundamental advances in artificial intelligence. One critical but challenging problem in AI is to demonstrate the ability to consistently understand and correctly answer general questions about the world. The aim of this challenge: using a dataset of multiple choice question and answers from a standardized 8th-grade science exam create a model that gets to the head of the class. Tools used: Recoll, Xapian, Python
Telstra Network Disruptions
February 1, 2016 – Present
81st place among 974 competitors (top 10%). Predict service faults on Australia's largest telecommunications network. Tools: XGBoost, t-SNE Github page: https://github.com/alexander-rakhlin/Telstra-Network-Disruptions
Rossmann Store Sales
December 1, 2015 – Present
275 place among 3303 teams (top 10%) In this competition, Rossmann is challenging you to predict 6 weeks of daily sales for 1,115 stores across Germany. Reliable sales forecasts enable store managers to create effective staff schedules that increase productivity and motivation. Tools: Python, Keras, XGBoost
Flavours of Physics: Finding τ → μμμ
October 1, 2015 – Present
5th place among 673 teams. Identify a rare decay phenomenon. The aim of this challenge is to find a phenomenon that is not already known to exist – charged lepton flavour violation – thereby helping to establish "new physics". Neural Networks, Gradient Boosted Trees, Python, Keras, XGBoost. My presentation for CERN workshop "Flavours of Physics Challenge: Physics Prize Transfer Learning approach" http://www.slideshare.net/AlexanderRakhlin/flavours-of-physics-challenge-transfer-learning-approach Github page: https://github.com/alexander-rakhlin/flavours-of-physics
Grasp-and-Lift EEG Detection
August 1, 2015 – Present
105th place among 379 competitors. Identify hand motions from EEG recordings. Tools and methods: Neural Networks. Python, Neon
Diabetic Retinopathy Detection Challenge
July 1, 2015 – Present
131 position among 661 teams (top 25%). With color fundus photography as input, built an automated system for Diabetic Retinopathy Detection. Tools and methods: Python, Theano, Keras Deep Learning library, AWS. See my article on LinkedIn: https://www.linkedin.com/pulse/machine-learning-diabetic-retinopathy-detection-alexander-rakhlin Github: https://github.com/alexander-rakhlin/kaggle_drd
Otto Group Product Classification Challenge
May 1, 2015 – Present
218 position among 3514 teams (top 10%). Build a predictive model which is able to distinguish between main product categories. Tools: Matlab, Python, Keras Deep Learning library, XGBoost Github: https://github.com/alexander-rakhlin/kaggle_otto
National Data Science Bowl
March 1, 2015 – Present
203 position among 1049 teams (top 25%) Build an algorithm to automate plankton image identification. Tools and methods: Matlab, Python, Caffe (Convolutional Neural Networks), GraphLab
American Epilepsy Society Seizure Prediction
November 1, 2014 – Present
245th position among 504 teams. Develop a system for forecasting seizures in dogs and humans with naturally occurring epilepsy. Tools and methods: Matlab, Python, Theano (CNNs).
DecMeg2014 - Decoding the Human Brain
July 1, 2014 – Present
Develop a system to predict visual stimuli from MEG recordings of human brain activity. Tools and methods: Matlab, Python
Algo trading
December 1, 2007 – Present
Algo trading. Research and development of market models and algorithms for the derivative market on Moscow Exchange. Implemented and put into successful work an automated system for trading futures and options on equities index. Languages used: Matlab/R/C#
Neural Networks for Machine Learning
Coursera
June 24, 2026 – Present
Data Visualization
Coursera
June 24, 2026 – Present
Beginning Game Programming with C#
Coursera
June 24, 2026 – Present
Introduction to Systematic Program Design
Coursera
June 24, 2026 – Present
General Game Playing
Coursera
June 24, 2026 – Present
Pattern-Oriented Software Architectures for Concurrent and Networked Software
Coursera
June 24, 2026 – Present
Natural Language Processing
Coursera
June 24, 2026 – Present
Probabilistic Graphical Models
Coursera
June 24, 2026 – Present
Statistical Learning
Stanford Online
June 24, 2026 – Present
CS1156x: Learning From Data
edX
June 24, 2026 – Present
Cluster Analysis in Data Mining
Coursera
June 24, 2026 – Present
Pattern Discovery in Data Mining
Coursera
June 24, 2026 – Present
Cultural Fit Analysis
The candidate's career trajectory shows a significant shift from finance/banking to machine learning and data science, primarily through self-employment and competition participation. While this demonstrates strong self-motivation and adaptability, the lack of traditional corporate data analyst roles might require a cultural adjustment. The projects are highly technical and research-oriented, aligning well with a data-driven culture. However, the target role 'Data Analyst' might be a slight mismatch for the candidate's deep expertise in advanced ML/DL, which typically aligns more with Data Scientist or ML Engineer roles. The breadth of skills is strong within the ML/DL domain, but less explicit on traditional data analyst tools like advanced SQL, BI tools, or data warehousing.
Soft Skills & Operational Fit
The candidate's extensive participation and high rankings in numerous data science competitions suggest strong problem-solving abilities, dedication, and a results-oriented approach. The self-employed period focusing on Machine Learning projects indicates a high degree of autonomy and initiative. However, there is no direct data to assess communication, stress handling, or team collaboration in a corporate setting.