Ivan Poleschyuk, Computer Vision Developer in Cape Town, Western Cape, South Africa
Ivan Poleschyuk

Computer Vision Developer in Cape Town, Western Cape, South Africa

Member since February 21, 2018
Ivan is a passionate machine learning engineer and full-stack software developer with a master's degree in computer science. His expertise includes machine learning and computer vision technologies, with proficiencies in Python and R in the data science field. Ivan is also experienced in leading and managing development teams.
Ivan is now available for hire




Cape Town, Western Cape, South Africa



Preferred Environment

Ubuntu, Git, PyCharm

The most amazing...

...thing I've built is an offline deep learning-based user identity verification workflow for a smartphone.


  • Computer Vision Developer

    2019 - PRESENT
    • Built a Python-based automatic video streaming pipeline for stadiums.
    • Implemented real-time multiple 4K cams panorama stitching.
    • Developed an automatic 'video operator' algorithm.
    • Implemented deep learning-based game situation understanding.
    • Set up a DevOps process with CI/CD and monitoring (Bitbucket, Docker, Jenkins, and ELK).
    • Prototyped algorithms with C++ (realtime high-res stitching, camera calibration, and lens dewarping).
    Technologies: Deep Learning, Computer Vision, Machine Learning, Jenkins, Docker, Bitbucket, C++, Python, Wowza, OpenCV, FFmpeg, PyTorch, Keras
  • Senior Video Research Engineer

    2020 - 2020
    • Researched video transcoding verification approaches.
    • Built a high-performance video transcoding verification API.
    • Developed a deep learning model to verify transcoded videos.
    • Explored attack vectors of zero-trust blockchain-based P2P video transcoding network.
    Technologies: Google Cloud Platform (GCP), FFmpeg, Blockchain, OpenCV, Python, Keras
  • Augmented Reality Developer

    2019 - 2019
    Lynx Equity Limited (via Toptal)
    • Built a high-performance CV pipeline for an embedded device.
    • Created automatic unit tests on real data.
    • Implemented a holographic recorder app based on Microsoft SDK.
    • Built a code to map data from several sensors to a single origin real-world coordinates.
    Technologies: C++, UWP, HoloLens
  • Data Scientist

    2018 - 2019
    Media Startup (via Toptal)
    • Developed a news data crawler.
    • Created a data processing pipeline.
    • Built a recommendation engine based on Elasticsearch.
    • Implemented an admin web UI for the recommendation engine.
    Technologies: Flask, Elasticsearch, Python
  • Machine Learning Expert

    2018 - 2019
    Alfa (via Toptal)
    • Developed an NLP processing pipeline for candidate resume analysis.
    • Consulted in-house development team on AI technologies.
    Technologies: Scikit-learn, XGBoost, NLTK, SpaCy, Python
  • CTO

    2018 - 2019
    Digital Identity Startup (via Toptal)
    • Built a prototype with Computer Vision and AI components running offline on mobile device (TensorFlow, MobileNet, OpenCV).
    • Established DevOps process and AWS infrastructure (Jira, Jenkins, Kanban, Docker).
    • Created an API and backend services with Flask.
    • Managed the project and the team, including hiring and interviews.
    • Created a client area of the company's website using Django.
    Technologies: Facial Recognition, Deep Learning, OpenCV, TensorFlow, iOS, Android, Swift, C++, Java, Django, Flask, Python
  • Team Leader, Lead Data Scientist

    2010 - 2017
    Institute of Information Systems, Inc.
    • Led a team developing ECM system for a local legislative assembly.
    • Built a reporting product with built-in analytics based on Microsoft SQL Server Reporting and Analysis services (OLAP).
    • Set up DevOps process in the department from scratch.
    • Implemented an anomaly detector for city's traffic monitoring system.
    Technologies: Ubuntu, Python, Microsoft
  • Software Developer

    2007 - 2010
    IT Business, Inc.
    • Worked on custom ECM for an industrial facility.
    • Built a batch scanning plugin for EMC Documentum and IBM Content Manager.
    • Developed a project synchronization tool for Microsoft Project Server.
    Technologies: EMC Documentum, JavaScript, Java, .NET


  • My Blog

    My blog about machine learning, computer vision, and other programming stuff.

  • Assessing the Quality of Innovative Medical Equipment

    I have analyzed the metrics produced by vertebral column scanning machine of novel architecture (3D scanning and motion capture based) with R. I had anonymous data of repetitive visits of more than 500 patients and used ggplot2 to visualize correlations between metrics for repetitive visits of the same patient versus data of unrelated patients. My report had outlined which metrics are most useful for patient diagnostics and which are too noisy or are unlikely to be linked to any medical issue.

  • Building a Trading Robot With Predictive Analytics

    The goal of the project was to build a model, which will produce an additional input to the trading system (along with technical indicators) built by a private client for a day trading on a MICEX futures market. I have used R (dplyr, ggplot2, neuralnet, betategarch, xts, ttr) to work on this task. The main challenge, as always with financial time series, was the dynamic nature of the process modeled. GARCH was used to estimate the optimal time window to train the model for predictions. It also turned out that simpler models, like logistic regression, converge better on such data.

  • Estimating Animals' Weights

    This amusing project was about estimating a weight of each of several small animals in the cage (exact number unknown) using data feed from one IoT scales in each cage. I used Python (pandas, scipy, numpy, matplotlib) to visualize the data and work along with the client on an optimal clustering-based algorithm to determine the exact weight of each animal. The main challenge was the fact that there was no way of telling if animal climbed on the scales entirely.

  • Urban Traffic Analysis

    In this project, the goal was to store and analyze the data from hundreds of CCTV traffic cameras across the city to detect rush-hour bottlenecks, count number of vehicles, detect speed limit violations. I've built PySpark scripts to collect these statistics over an Apache Spark cluster. The data was stored in CSV format across multiple Hadoop nodes, several hundred gigabytes in total.

  • Music Recommendation Service

    This is the music recommendation service I've built as a hobby project exclusively on open source technologies.

  • Kaggle Competition: Trip Type Classification

    59th (top 6%) - Walmart Recruiting: Trip Type Classification. This competition involved complex market basket analysis. I have used R to preprocess (dplyr) and explore the data (ggplot2), then run a forecast with random forest and xgboost, and fed preprocessed dataset to neural net model built with Python to ensemble predictions.

  • ArthroLens: Augmented Reality for Operating Room

    In this project, I have worked as one of computer vision developers to build a Microsoft HoloLens augmented reality application to assist the surgeon in a knee replacement surgery. The goal was to eliminate some of the hospital's costs associated with buying expensive guiding devices and replace them with holograms. Main technologies: C++, OpenCV, UWP, and HoloLens.

  • Zero to Hero: Flask Production Recipes (Publication)
    Flask is a great way to get up and running quickly with a Python applications, but what if you wanted to make something a bit more robust? In this article, Toptal Freelance Python Developer Ivan PoleschyuI shares some tips and useful recipes for building a complete production-ready Flask application.


  • Languages

    Python, C#, SQL, R, C++, Java, Swift, JavaScript
  • Frameworks

    Flask, Django, .NET
  • Libraries/APIs

    OpenCV, Scikit-learn, TensorFlow, SQLAlchemy, FFmpeg, PyTorch, Keras, NLTK, SpaCy, XGBoost
  • Tools

    PyCharm, Git, GitHub, GitLab, Celery, Amazon SQS, HoloLens, Wowza, Bitbucket, Jenkins, ELK (Elastic Stack), RabbitMQ
  • Platforms

    Jupyter Notebook, Ubuntu, Windows, Linux, Amazon Web Services (AWS), Amazon EC2 (Amazon Elastic Compute Cloud), Microsoft, Android, iOS, UWP, Docker, Blockchain, Google Cloud Platform (GCP), RStudio, CUDA
  • Storage

    PostgreSQL, Elasticsearch, SQL Server Analysis Services (SSAS), SQL Server Reporting Services (SSRS), SQL Server 2010, MySQL, Amazon S3 (AWS S3)
  • Other

    Artificial Intelligence (AI), Artificial Neural Networks (ANN), Video Streaming, Video Processing, Machine Learning, Recommendation Systems, Neural Networks, Computer Science, Computer Vision, Deep Neural Networks, Deep Learning, Facial Recognition, Tesseract, Convolutional Neural Networks, Image Analysis, Time Series Analysis, Gunicorn, Natural Language Processing (NLP), Speech to Text, Speech Recognition, OCR, Augmented Reality (AR), Multidimensional Expressions (MDX), EMC Documentum, Statistics
  • Paradigms

    Agile Software Development, OLAP


  • Master's Degree in Computer Science, Mathematics
    2004 - 2010
    Moscow State University - Moscow, Russia


  • Udacity Self-driving Car Engineer Nanodegree
  • Microsoft Certified Solutions Developer: App Builder
  • Machine Learning
    APRIL 2016 - PRESENT
  • Microsoft Certified Solutions Developer: Web Applications
  • Microsoft Specialist: Programming in HTML5 with JavaScript and CSS3

To view more profiles

Join Toptal
Share it with others