Vahagn Altunyan, Developer in Yerevan, Armenia
Vahagn is available for hire
Hire Vahagn

Vahagn Altunyan

Verified Expert  in Engineering

Machine Learning Engineer and Julia Developer

Location
Yerevan, Armenia
Toptal Member Since
June 10, 2021

Vahagn is a Julia developer and machine learning engineer with substantial experience in competitive programming and mathematical competitions. He has won gold and two silver medals from the IMC math competition. Solving hard algorithmic problems is his hobby, and serves his clients well on complex projects. He has created a competitive programming problem-solving platform, a source code testing system, and a data collection and annotation system. Vahagn also enjoys developing APIs.

Availability

Part-time

Preferred Environment

Windows, Visual Studio Code (VS Code), Visual Studio 2019, Windows Subsystem for Linux (WSL)

The most amazing...

...and unexpected result I delivered was in a math competition, where I proved the theorem of Darboux without knowing about it and solved the challenge.

Work Experience

Computer Vision Engineer

2017 - 2020
Improvis
  • Architected and developed a back end for a training data collection and annotation system.
  • Developed a 2-role application for runners and coaches. Each runner has a calendar with a training plan, and a coach can be assigned to athletes to manage his/her calendar.
  • Built an application where users upload a video performing an exercise, and the poses of the user in frames are detected with a pose estimation algorithm (OpenPose). Based on the positions, the algorithm then gives a medical diagnoses.
Technologies: OpenCV, Python 3, Django, TensorFlow, PyTorch, Qt 5, C++

Software Engineer

2018 - 2018
Freelancing
  • Developed a source code testing system with test cases in an isolated environment for competitive programming problems measuring memory usage and running time.
  • Implemented a code editor and a generic syntax highlighting system for programming languages using QT/C++.
  • Created a competitive programming problem-solving platform where users can solve exciting problems and submit solutions for them. The solution is tested in an isolated environment on the server, with results displayed in the front end.
Technologies: Qt 5, C++, Python 3, REST, HTML, CSS

Data Collection System

Data collection and annotation system that allows users with data uploader roles to upload videos or a collection of images. Data annotators then take images pending for annotation and annotate them for semantic segmentation and mark tags and descriptors for each image collection. In the end, a user with a data consumer role can filter out data with tags and other descriptors and download images with COCO-style annotations.

Judge System

A judging system for testing user solutions against a set of test cases.
Administrators can create and upload problems and test cases (including checker algorithms) and specify resource limitations for the correct solution. Regular users can sign up into the system and start solving already uploaded problems. They can choose the programming language for their solution and write the integrated editor's code with a generic syntax highlighting system. Submitted solutions are tested against the prepared test cases in an isolated environment, and the result of the evaluation is shown to the user.

Languages

C++, Python, Julia, Python 3, HTML, CSS, HTML5, CSS3

Frameworks

Django, Django REST Framework, Qt 5, Flask

Libraries/APIs

OpenCV, TensorFlow, PyTorch

Paradigms

REST, Data Science

Platforms

Windows, Visual Studio Code (VS Code)

Other

Probability Theory, Data Structures, Algorithms, Mathematics, Linear Algebra, Neural Networks, Convolutional Neural Networks (CNN), Windows Subsystem for Linux (WSL), Image Processing

Tools

Git, Visual Studio 2019, GitHub

2020 - 2021

Master's Degree in Computer Science

Yerevan State University - Yerevan, Armenia

2016 - 2020

Bachelor's Degree in Computer Science

Yerevan State University - Yerevan, Armenia

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.

1

Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3

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