Ales Fexa

About me

I am an AI software engineer from Prague with a special passion for computer vision, LLMs, and mathematics.

I spend my workdays writing various Python and .NET applications. In my spare time, I develop Sefexa, a free tool for semi-automatic image segmentation.

My CV and publications are on this site, detailed information about my skills and career can be found on LinkedIn.

CV

Education:

November 2004 – February 2008:

Institut für Algorithmen und Kognitive Systeme, Universität Karlsruhe (TH), Germany

"Doktor der Ingenieurwissenschaften" (PhD) in Computer Science

October 1999 – July 2002, January 2004 – September 2004:

Faculty of Mathematics and Physics, Charles University in Prague

"Magistr" (Master of Science) in Computer Science (specialization in Computer Vision, Artificial Neural Networks, Artificial Intelligence)

January 2003 – December 2003:

Kansas State University, Manhattan, Kansas, USA

Master of Science in Computer Science, GPA 4.0/4.0

Languages:

German C2

English C2

Czech (native language)

Work experience:

Since November 2025:

AI Software Engineer (Python/.NET), Independent Contractor, remote from Nýdek, Czechia

Development of AI systems based on LLMs and computer vision, including end-to-end model design, implementation, and evaluation.

September 2016 - October 2025:

Software Engineer / Data Scientist (.NET/Python) at Enverus, remote from Nýdek, Czechia

Design and implementation of natural language processing algorithms for text data extraction (n-grams, rules, LLM prompts etc.) to process millions of documents daily through cloud-based computation.

March 2012 - July 2016:

Software Engineer (.NET) at Barclays, Prague, Czechia

Development of front-end (Silverlight) and back-end (C# and MsSql) applications for traders.

March 2008 - February 2012:

Software Engineer (.NET) at T-SOFT, a.s., Prague, Czechia

Web applications, real time visualization of infrared video, size estimation of a cloud created by an explosion.

November 2004 – February 2008:

Graduate Research Assistant at Institut für Algorithmen und Kognitive Systeme, Universität Karlsruhe (TH), Germany

Research and implementation of computer vision algorithms, knowledge representations, and natural language generation (Java).

March 2004 – July 2004:

Software Developer (C/C++) at Neovision s.r.o., Prague

Design of algorithms for automatic translation of Braille texts.

July 2002 – December 2002:

Undergraduate Research Assistant at Fraunhofer Institut Medienkommunikation, Schloss Birlinghoven, Germany

Research and implementation of algorithms for image segmentation (C/C++).

Skills currently used:

C# ( years)

Python (5 years)

MsSql (10 years)

PySpark (5 years)

GitHub (12 years)

AWS (4 years)

Pandas (2 years)

LLM prompts (1 years)

Kafka (1 year)

Elasticsearch (1 year)

Docker (7 years)

Jira (14 years)

Skills used in the past:

ASP.NET (3 years)

Java (3 years)

Hobbies:

Computer Vision (mainly Image Segmentation)

Mathematics (mainly in the context of Computer Vision)

Blackberries (in our garden)

Publications

2008

A. Fexa: Erzeugung mehrsprachlicher Beschreibungen von Verkehrsbildfolgen. Dissertation, Fakultät für Informatik der Universität Karlsruhe (TH), August 2008; Dissertationen zur Künstlichen Intelligenz (DISKI) 317; Akademische Verlagsgesellschaft AKA GmbH, Berlin, 2008.

2007

A. Fexa and H.-H. Nagel: Integrating Behavior-based Prediction for Tracking Vehicles in Traffic Videos. In: Proc. of the International Conference on Computer Vision Systems (ICVS2007), 21-24 März 2007, Bielefeld, Germany. Full Article.

2006

A. Fexa, K. Uchimoto, H. Isahara, J. Kawai, and H.-H. Nagel: On 'Speaking-Cameras'. In: Ch. Freksa, M. Kohlhase, and K. Schill (Eds.): Proc. of the 29th Annual German Conference on Articial Intelligence (KI 2006), Poster Abstracts, June 2006, Bremen, pp. 5-7. Full Article.

M. Kolesnik, A. Fexa: How robust is the SVM wound segmentation? In: Proc. of the 7th Nordic Signal Processing Symposium (NORSIG'2006), June 7-9, 2006, Reykjavik, Iceland, IEEE ISBN 1-4244-0413-4/06, pp. 50-53. Full Article (©Springer-Verlag).

A. Fexa, H.-H. Nagel, A. Ottlik, and J. Siecke: Behavior-based 3D-Tracking During Unmodelled Occlusion. In: M. Vincze and L. Paletta (Eds.): Proc. of the 2nd International Cognitive Vision Workshop (in conjunction with ECCV-2006), 13 May 2006, Graz, Austria, pp. 1-8 (on the Workshop CD).

2005

M. Kolesnik, A. Fexa: Multi-dimensional Color Histograms for Segmentation of Wounds in Images. In: Proc. of the International Conference on Image Analysis and Recognition (ICIAR'2005), Toronto, Canada, September 28-30, 2005, Lecture Notes in Computer Science, Volume 3656/2005, pp. 1014-1022. Full Article (©Springer-Verlag).

A. Fexa: Dependence of Conceptual Representations for Temporal Developments in Videosequences on a Target Language. In: D. Paulus and D. Droege (Eds.): Proc. of the KI 2005 Workshop 7 'Mixed-reality as a Challenge to Image Understanding and Articial Intelligence', 11 September 2005, Koblenz; Fachberichte Informatik, Universität Koblenz-Landau, Institut für Informatik, Universitätsstr. 1, 56070 Koblenz, pp. 47-54.

2004

M. Kolesnik, A. Fexa: Segmentation of Wounds in the Combined Color-Texture Feature Space. In: Proc. of SPIE: Medical Imaging 2004, San-Diego, USA, 16-19 February 2004, Volume 5370, pp. 549-556. Full Article (©Springer-Verlag).