Iaroslav Khripkov Email & Phone Number

Москва, Москва, Россия

• Training university students to successful participation in ICPC
• Led the trained team to the ICPC World semifinals
• Prepared course program including lectures, seminars, hometasks, slides e.t.c
• Explained complicated algorithmic problems and approaches to students without any Olympiad background
• Organized online testing system and it’s integration with recorded lectures on YouTube

Moscow, Russia

Redesigned and optimized data processing pipeline. Gained 3x+ acceleration on long samples of text.Created and developed entire raw data processing system with parallel computing and distributed systems which utilized over 3 TB of raw data.Refactored CI/CD pipelines, entire kubernetes cluster task configs and Docker images for a large codebase engineers.

Moscow, Russia

* Designed, programmed and trained a large-scale multimodal (CV+NLP) neural network for content similarity analysis, utilizing hundreds of terabytes of data (including images, text, and speech) for improved accuracy and better insights. * Devised and implemented a highly efficient template matching algorithm to detect content across multiple platforms..

Moscow, Moscow City, Russia

• Developed and improved DL model for splitting a video into fragments within common theme. Achieved quality about 90% of human performance.
• Reduced data extraction runtime by migrating from python program to C++ approach for Map-Reduce system. Runtime reduced by 86%.
• Created pipeline for feature extraction with computer vision model for face recognition.
• Implemented face-based index for reverse video search with AkNN.

Москва, Россия

• Operated Internet users recognition system based on their hardware specs and digital habits
• Improved anomaly detection model for bots and fraud detection
• Engineered explanatory AI, using gradient boosting model (CatBoost) to generate understandable rules for decision making
• Developed RNN for analysing digital fingerprints to highly precise frauders and bots detection
• Processed huge amounts of data by using ClickHouse, YQL (SQL dialect), powerful clusters
• Optimized data transfer between different data clusters and databases by implementing automated pipeline written on python and bash

Москва, Москва, Россия

• Utilized a ML approach to develop a precise thyroid disease classifier that achieved an accuracy rate of 99.9
• Conducted numerous experiments with a wide range of ML models to build polynomial regression and time series prediction models, resultingin successful stock volatility predictions.
• Designed a chatbot that predicts diagnoses based on user complaints in free text, using a clusterization model that groups diagnoses andcomplaints by their context. Achieved an impressive accuracy score of 0.94 by.

Master Of Science - Ms, Computer Science

Part-time MSc in Software Engineering

Master Of Science - Ms, Mathematics And Computer Science

Master Of Science - Ms, Machine Learning

• C++ (Designed from scratch such projects as ray-tracer, LISP Interpreter, Telegram Bot using POCO library) • Mathematics for Data.

Bachelor Of Science - Bs, Mathematics And Computer Science

• Introduction to programming, Informatics courses (C++ in both) • Object Oriented Programming course (Java). Projects: Fractal plotter.