Gleb Tkachev

ML Scientist

Amazon Alexa

About me

I am an Applied ML Scientist at Amazon Alexa, working on NLP and Question Answering. Prior to that in my PhD I worked on ML for rendering and analysis of scientific data. I focused on self-supervised methods and had occasional encounters with generative image models, style transfer and reinforcement learning. Additionally, I have solid software development skills and 3+ years of industry experience.

Featured Skills

Machine Learning
- Self-Supervised Learning
- Autoencoders
- NLP
Software Development
- Python (Pytorch, Tensorflow, Numpy)
- C++

Education

PhD Computer Science, 2022

University of Stuttgart
M.Sc. Computer Science, 2017

University of Stuttgart
B.Sc. Automatic Systems, 2014

Kaliningrad Technical

Featured Publications

S4: Self-Supervised learning of Spatiotemporal Similarity

Gleb Tkachev, Steffen Frey, Thomas Ertl

August, 2021 TVCG

S4: Self-Supervised learning of Spatiotemporal Similarity

Metaphorical Visualization: Mapping Data to Familiar Concepts

Gleb Tkachev, Rene Cutura, Michael Sedlmair, Steffen Frey, Thomas Ertl

April, 2022 CHI 2022 Extended Abstracts (alt.CHI)

Metaphorical Visualization: Mapping Data to Familiar Concepts

See all publications

Experience

Applied ML Scientist

Amazon Alexa

Oct 2022 – Present 1 m Berlin, Germany

Working on NLP, Question Answering and Ranking.

PhD Researcher

University of Stuttgart

Feb 2017 – Present 5 y 8 m Stuttgart, Germany

Researching Machine Learning approaches to visualization and analysis of scientific data (see publications). Teaching practical Master courses and supervising Master theses.

Scientific Visualization
Self-Supervised Learning
Convolutional Networks
Human + ML
High-Performance Computing
Python (Numpy, Pytorch, Keras, Tensorflow)
C++ (MPI, CUDA)

Web Developer

Bett Ingenieure

Apr 2015 – Dec 2016 1 y 8 m Stuttgart, Germany

A part-time web-development job during my Master studies (20h/week).

Created web-based business applications and development tools. (PHP + CSS + Javascript)
Linux administration. (Debian + Apache + Freeswitch)

Research Assistant

University of Stuttgart

Feb 2015 – May 2015 3 m Stuttgart, Germany

A part-time web-development job during my Master studies (20h/week).

Developed an information visualization prototype. (Javascript + D3)

Game Developer

AIGRIND

Sep 2013 – Sep 2014 1 y Kaliningrad, Russia

Mobile MMORPG server development. (C# + .NET)
Created a compiler for an in-house networking solution. (C# + .NET + Irony)
Mobile game development, rapid prototyping. (Unity3D)

Web Developer

INOK Group

Jul 2012 – Dec 2012 6 m Kaliningrad, Russia

Web-based business application development. (PHP + Yii, PostgreSQL, Javascript + Knockout)
Web-server administration. (Ubuntu server, Apache, Postfix, Samba)

Publications

Gleb Tkachev, Rene Cutura, Michael Sedlmair, Steffen Frey, Thomas Ertl (2022). Metaphorical Visualization: Mapping Data to Familiar Concepts. CHI 2022 Extended Abstracts (alt.CHI).

PDF Cite Video DOI Talk Appendix

Hamid Gadirov, Gleb Tkachev, Thomas Ertl, Steffen Frey (2021). Evaluation and Selection of Autoencoders for Expressive Dimensionality Reduction of Spatial Ensembles. ISVC 2021.

PDF Cite DOI Appendix

Moritz Heinemann, Steffen Frey, Gleb Tkachev, Alexander Straub, Filip Sadlo, Thomas Ertl (2021). Visual Analysis of Droplet Dynamics in Large-Scale Multiphase Spray Simulations. JVis.

PDF Cite DOI

Gleb Tkachev, Steffen Frey, Thomas Ertl (2021). S4: Self-Supervised learning of Spatiotemporal Similarity. TVCG.

PDF Cite Code Video DOI Appendix Talk

Shivam Agarwal, Gleb Tkachev, Michel Wermelinger, Fabian Beck (2020). Visualizing Sets and Changes in Membership Using Layered Set Intersection Graphs. VMV2020.

PDF Cite DOI

Gleb Tkachev, Steffen Frey, Thomas Ertl (2019). Local Prediction Models for Spatiotemporal Volume Visualization. TVCG.

PDF Cite DOI

Ilyass Tabiai, Gleb Tkachev, Patrick Diehl, Steffen Frey, Thomas Ertl, Daniel Therriault, Martin Lévesque (2019). Hybrid Image Processing Approach for Autonomous Crack Area Detection and Tracking Using Local Digital Image Correlation Results Applied to Single-Fiber Interfacial Debonding. Engineering Fracture Mechanics.

PDF Cite Code Dataset DOI

Gleb Tkachev, Steffen Frey, Christop Müller, Valentin Bruder, Thomas Ertl (2017). Prediction of Distributed Volume Visualization Performance to Support Render Hardware Acquisition. EGPGV 2017.

PDF Cite DOI