Proof of concept Implementierung für die Integration von IOTA Flash Channels in Hausautomatisierungssysteme (z.B. Home Assistant).
GitHub: [URL on request]
Awards: [URL on request]
The novel, federated approach for Industry B2B Cloud Platforms.
GitHub: [URL on request]
Email basierte Wallet für Stellar Lumen die es ermöglicht einfach Lumen zwischen Emails zu tauschen.
GitHub: [URL on request]
Speeches in the National Council of Austria are stenographically recorded and made available on data.gv.at. Recorded texts were used to train an artificial recurrent neural network, which uses the last 15 words to predict each 16th word. The source code is available on GitHub.
GitHub: [URL on request]
Website: [URL on request]
2016 till 2018
Customer:Salzburg Research
Role:IoT Researcher with Machine Learning focus.
Tasks:
Since the summer of 2016, I am working at Salzburg Research in the area of Industrial
Internet of Things. During this time, I dealt with the measurement of 3D printing processes
and the subsequent analysis of the measured data. The goal was to predict the failure of
individual hardware components in order to carry out predictive maintenance. In another
project, I looked at a cloud platform that will enable small and medium-sized businesses
to find new partners on the platform and augment advanced business processes.
2014 - 2016
Customer:CrossCloud GmbH
Role:Co-Founder, CTO and Managing Director
Tasks:
CrossCloud is a client application for cloud storage, e.g. Dropbox or Google Drive, which
enables companies to use multiple cloud storage services in a very easy and secure way.
As co-founder and CTO of a startup I led a team of five developers and learned how to
create a product from the ground up. During my role as scrum master I learned how to
use resources efficiently. The founders of CrossCloud received the award of being Grazer
of the year 2015 in the field of economy. Two financial rounds were closed and a FFG
Basisprogramm was approved successfully.
2012 - 2014
Role:Freelancer, App and Web Development.
Tasks:
As a freelancer I developed several Groovy on Rails and HTML 5 applications.
2012 - 2013
Role:Teaching Assistent
Customer: Graz University of Technology.
Tasks:
Institute of Electrical Measurement and Measurement Signal Processing (EMT)
In the two years I was guiding students at the Image Based Measurement Labor.
2008 - 2009
Military Service
Schwarzenbergkaserne, Siezenheim.
IT Projects
Python,TensorFlow,Keras
Neural Politician.
Speeches in the National Council of Austria are stenographically recorded and made
available on data.gv.at. Recorded texts were used to train an artificial recurrent neural
network, which uses the last 15 words to predict each 16th word. The source code used is
available on
Python,TensorFlow,Keras
Deep Learning Trading Agent.
With the help of a Convolutional Neural Nets (CNNs), we tried to make price predictions
of cryptocurrencies (e.g. Bitcoin or Ethereum). The neural network was implemented with
Google?s Tensorflow and the higher abstraction library Keras. Past trades of the different
currencies since 2013 were used as underlying dataset. Later, a sentiment analysis of
articles on Twitter was applied to further improve the prediction capabilities of the model.
Next steps in this project will be a semantic classification of texts using Word Embeddings
and an evaluation of Recurrent Neural Networks (RNN) for time series modelling.
Java, Groovy,Python, NodeJS
CrossCloud Client Application.
The first version of the client application was a desktop application written in Java
and Groovy. Connections to different cloud storage services and the synchronisation
were executed on the Java Virtual Machine. By the end of 2015 we rewrote the entire
application in Python. Mobile applications were developed using the Xamarin platform
in order to reuse the business logic on iOS and Android. Server-side implementation was
created using the Flask microframework.
Java, Python,TensorFlow
Cloud Storage Performance Analysis.
My master thesis dealt with a global performance analysis of three of the most used
cloud storage providers (i.e. Dropbox, Google Drive and OneDrive). Different regression
models were used to find an optimal model of global bandwidths. A deep understanding
of Multiple Linear Regression and Support Vector Machines has been developed within
this project.
Visual C++
Auto-routing Algorithmus.
During summer of 2011 and 2012 my task was to develop an auto-routing algorithm for
the the company Sigmatek GmbH in order to find the optimal path of electrical leads on
a board.
C, Assembler
Safe Implementation of the AES-algoritm.
During my bachelor thesis I was securing an existing implementation of the Advanced
Encryption Standard (AES) on an 8-bit microcontroller against side-channel attacks.
Computational Intelligence
Computer Vision
Master’s thesis: Cloud Storage Performance Analysis
Safe Implementation of AES
Libraries
OpenCV / Tensorflow / TensorFlow Lite / Keras / scikit-learn / Numpy / Flask
/ Electron / NodeJS / ReactJS / React Native / Django
Software
iPython Notebook / PyCharm / IntelliJ / Xcode / Latex Machine Learning / Artificial Intelligence
Regression
Multiple Linear Regression / Gaussian Processess (GP) / Support Vector Regression
(SVR)/ Neural Networks
Classification Support
Vector Machines (SVM)/ K-Nearest Neighbor (KNN) / Logistic Regression/ Neural Networks / Decision Trees
Clustering
K-Means / Gaussian Mixture Models (GMM) / Support Vector Clustering (SVC)/ Principal Component Analysis (PCA)
Deep Learning
Recurrent Neural Network (RNN) / Convolutional Neural Network (CNN) / Cross Entropy Error / Autoencoder
Proof of concept Implementierung für die Integration von IOTA Flash Channels in Hausautomatisierungssysteme (z.B. Home Assistant).
GitHub: [URL on request]
Awards: [URL on request]
The novel, federated approach for Industry B2B Cloud Platforms.
GitHub: [URL on request]
Email basierte Wallet für Stellar Lumen die es ermöglicht einfach Lumen zwischen Emails zu tauschen.
GitHub: [URL on request]
Speeches in the National Council of Austria are stenographically recorded and made available on data.gv.at. Recorded texts were used to train an artificial recurrent neural network, which uses the last 15 words to predict each 16th word. The source code is available on GitHub.
GitHub: [URL on request]
Website: [URL on request]
2016 till 2018
Customer:Salzburg Research
Role:IoT Researcher with Machine Learning focus.
Tasks:
Since the summer of 2016, I am working at Salzburg Research in the area of Industrial
Internet of Things. During this time, I dealt with the measurement of 3D printing processes
and the subsequent analysis of the measured data. The goal was to predict the failure of
individual hardware components in order to carry out predictive maintenance. In another
project, I looked at a cloud platform that will enable small and medium-sized businesses
to find new partners on the platform and augment advanced business processes.
2014 - 2016
Customer:CrossCloud GmbH
Role:Co-Founder, CTO and Managing Director
Tasks:
CrossCloud is a client application for cloud storage, e.g. Dropbox or Google Drive, which
enables companies to use multiple cloud storage services in a very easy and secure way.
As co-founder and CTO of a startup I led a team of five developers and learned how to
create a product from the ground up. During my role as scrum master I learned how to
use resources efficiently. The founders of CrossCloud received the award of being Grazer
of the year 2015 in the field of economy. Two financial rounds were closed and a FFG
Basisprogramm was approved successfully.
2012 - 2014
Role:Freelancer, App and Web Development.
Tasks:
As a freelancer I developed several Groovy on Rails and HTML 5 applications.
2012 - 2013
Role:Teaching Assistent
Customer: Graz University of Technology.
Tasks:
Institute of Electrical Measurement and Measurement Signal Processing (EMT)
In the two years I was guiding students at the Image Based Measurement Labor.
2008 - 2009
Military Service
Schwarzenbergkaserne, Siezenheim.
IT Projects
Python,TensorFlow,Keras
Neural Politician.
Speeches in the National Council of Austria are stenographically recorded and made
available on data.gv.at. Recorded texts were used to train an artificial recurrent neural
network, which uses the last 15 words to predict each 16th word. The source code used is
available on
Python,TensorFlow,Keras
Deep Learning Trading Agent.
With the help of a Convolutional Neural Nets (CNNs), we tried to make price predictions
of cryptocurrencies (e.g. Bitcoin or Ethereum). The neural network was implemented with
Google?s Tensorflow and the higher abstraction library Keras. Past trades of the different
currencies since 2013 were used as underlying dataset. Later, a sentiment analysis of
articles on Twitter was applied to further improve the prediction capabilities of the model.
Next steps in this project will be a semantic classification of texts using Word Embeddings
and an evaluation of Recurrent Neural Networks (RNN) for time series modelling.
Java, Groovy,Python, NodeJS
CrossCloud Client Application.
The first version of the client application was a desktop application written in Java
and Groovy. Connections to different cloud storage services and the synchronisation
were executed on the Java Virtual Machine. By the end of 2015 we rewrote the entire
application in Python. Mobile applications were developed using the Xamarin platform
in order to reuse the business logic on iOS and Android. Server-side implementation was
created using the Flask microframework.
Java, Python,TensorFlow
Cloud Storage Performance Analysis.
My master thesis dealt with a global performance analysis of three of the most used
cloud storage providers (i.e. Dropbox, Google Drive and OneDrive). Different regression
models were used to find an optimal model of global bandwidths. A deep understanding
of Multiple Linear Regression and Support Vector Machines has been developed within
this project.
Visual C++
Auto-routing Algorithmus.
During summer of 2011 and 2012 my task was to develop an auto-routing algorithm for
the the company Sigmatek GmbH in order to find the optimal path of electrical leads on
a board.
C, Assembler
Safe Implementation of the AES-algoritm.
During my bachelor thesis I was securing an existing implementation of the Advanced
Encryption Standard (AES) on an 8-bit microcontroller against side-channel attacks.
Computational Intelligence
Computer Vision
Master’s thesis: Cloud Storage Performance Analysis
Safe Implementation of AES
Libraries
OpenCV / Tensorflow / TensorFlow Lite / Keras / scikit-learn / Numpy / Flask
/ Electron / NodeJS / ReactJS / React Native / Django
Software
iPython Notebook / PyCharm / IntelliJ / Xcode / Latex Machine Learning / Artificial Intelligence
Regression
Multiple Linear Regression / Gaussian Processess (GP) / Support Vector Regression
(SVR)/ Neural Networks
Classification Support
Vector Machines (SVM)/ K-Nearest Neighbor (KNN) / Logistic Regression/ Neural Networks / Decision Trees
Clustering
K-Means / Gaussian Mixture Models (GMM) / Support Vector Clustering (SVC)/ Principal Component Analysis (PCA)
Deep Learning
Recurrent Neural Network (RNN) / Convolutional Neural Network (CNN) / Cross Entropy Error / Autoencoder
Direktester geht's nicht! Ganz einfach Freelancer finden und direkt Kontakt aufnehmen.