Alexander is available for hire

Alexander Rhatushnyak

Verified Expert in Engineering

Data Compression Algorithms Developer

Location

Waterloo, ON, Canada

Toptal Member Since

March 13, 2018

Alexander possesses a Ph.D. in computer science, is an expert in lossless data compression algorithms, a Hutter Prize winner, and a top contributor in competitions that rank lossless data compression programs. His knowledge and experience also spans many adjacent fields including data science, machine learning, natural language processing, digital signing processing, and video/audio compression.

Algorithms C Data Mining Machine Learning ARM C++GCC Microsoft Visual Studio Python NumPy Boost Natural Language Processing (NLP)MATLAB Java SciPy

Portfolio

V-Nova

OpenCL, Python, C++, NVIDIA CUDA

Microsoft

SQL, C++

Experience

Data Compression Algorithms - 20 years Data Structures - 19 years C - 19 years Algorithms - 19 years C++ - 15 years Python - 6 years NumPy - 3 years Scikit-learn - 2 years

Availability

Part-time

Preferred Environment

Visual Studio, Windows, NumPy, Python, GCC, Ubuntu

The most amazing...

...things I've worked on are the video codecs for V-Nova (V-nova.com), a sentiment analysis for Sortable, and lossless image/text codecs listed in LPCB & LTCB.

Work Experience

Senior Software Engineer

2014 - 2018

V-Nova

Implemented advanced video compression for the modern web; this included some modern machine learning.
Wrote a reference video codec for a CPU; written in C++ and mostly done by myself.
Contributed to Cuda and OpenCL video codecs targeting GPU.
Researched and developed video compression algorithms; first with C++, then mostly with Python and NumPy, SymPy, and SciPy. Some of the R&D code was written in MATLAB. The telemetry data compression was again done with C/C++.
Developed lossless compression algorithms for medical images.

Technologies: OpenCL, Python, C++, NVIDIA CUDA

Software Development Engineer 2

2013 - 2014

Microsoft

Worked on the SQL Server debugging-only team.
Implemented various debugging methods; using C++ and some C#.

Technologies: SQL, C++

Experience

Hutter Prize Entry

http://prize.hutter1.net

I wrote this compressor, mostly in 2017, and it compresses 100 MB of English Wikipedia articles better than any other compressor. I was awarded the Hutter Prize on January 31, 2018.

The NLP half of compressor is a thousand times faster (when decompressing data) than the part with data modeling and machine learning methods, but it is equally important.

The version of this compressor modified to handle 1 GB of English Wikipedia data and is currently (as of Feb. 2018) at the very top of Large Text Compression Benchmark's rankings.

Competitions that Rank Lossless Data Compression Programs

I've participated in several competitions that rank lossless data compression programs, notably "Large Text Compression Benchmark" and "Lossless Photo Compression Benchmark". Please search for phda9 in "Large Text Compression Benchmark" and GraLIC, FLIC, QLIC, QIC in "Lossless Photo Compression Benchmark".

Skills

Languages

C, C++, Python, SQL, Java

Paradigms

Functional Programming, Object-oriented Programming (OOP), Data Science

Other

Algorithms, Data Structures, Data Compression, Data Compression Algorithms, Video Compression, ARM, Brainstorming, Data Mining, Machine Learning, OpenCL/GPU, Natural Language Processing (NLP), GPT, Generative Pre-trained Transformers (GPT)

Frameworks

Boost, OpenCL

Libraries/APIs

Scikit-learn, NumPy, SciPy

Tools

Microsoft Visual Studio, GCC, Visual Studio, MATLAB

Platforms

Ubuntu, Windows, Google Cloud Platform (GCP), NVIDIA CUDA

Storage

SQLite

Education

1999 - 2003

Ph.D. Degree in Computer Science

A.P. Ershov Institute of Informatics Systems - Novosibirsk, Siberia, Russia

Collaboration That Works

How to Work with Toptal

Toptal matches you directly with global industry experts from our network in hours—not weeks or months.

Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.

Start your risk-free talent trial

Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring