Keepoala – green loyalty program

2020/07 – now | full time project

Keepoala is the green loyalty program that allows customers to collect points for not returning whatever they buy online. The IT infrastructure is kind of complex and includes

  • Phone App for iOS and Android: React Native, Firebase Auth, Firebase FireStore
  • WebApp: React, FirebaseFirestore, AWS ECS + AWS ECR
  • Dashboard: R-Shiny, AWS PostGres
  • Shopify Plugin: React, NextJS, Firebase FireStore, Firebase Cloudfunctions
  • Shopware Pluigin: PHP
  • Shopware 6 Plugin: React, NextJS,Firebase Cloudfunctions
  • WooCommerce Plugin: PHP
  • PlentyMarkets Plugin: PHP
  • Billbee Connector: FirebaseCloudfunctions, React, AWS ECS + AWS ECR
  • additional Shop systems: LightSpeed, SalesForce CommerceCloud, Squarespace, Magento 2, OXID eSales

rayshaderanimate – Animate GPS data in 3D

2019/11 | spare time project

Rayshaderanimate allows using an .gpx file from any cycling or hiking trip to be animated as a video. Not only is the trip animated, but it is brought to a 3D landscape.

code: https://github.com/zappingseb/rayshaderanimate

read more: blog entry

Tools: R, ImageMagick, FFMPEG

bioWARP – The largest shiny app in the world

2017/09 – 2019/01 | Employee at Roche, Penzberg, Germany

bioWARP is an R shiny app used to analyze clinical trials at Roche Diagnostics. It is also used in maintenance and production processes. We called it the largest shiny app in the world as it consists of >500.000 lines of code. More to be found in my presentation: https://zappingseb.github.io/RPharma2018/

Webinar with RStudio: https://resources.rstudio.com/webinars/the-role-of-r-in-drug-discovery-research-and-development

Read more: blog entry

Tools: R, R-shiny, Selenium, docker, Jenkins, RHEL7-server, CSS

A leaflet Strava runner

2019/08 – spare time project

As a tiny project I analyzed the Strava data of a friend of mine. She is really fit and ran 5000 kilometers in around 5 years. I visualized her run in a mobile website where she really runs as a comic icon on a world map.

Website: http://engel-wolf.com/documents/elisah

Tools: HTML, js, leaflet

Stravachaser

2019/04 – spare time project

Stravachaser allows visualizing Strava segment data from different cities. Through this data it is possible to see in which city there are the fastest cyclists. The project is hosted here:

website: shinyapps.io

code: github.com

Tools: R, R-shiny, curl (scraping data), Selenium, CSS

BioPharma Visualization

Cluster network visualization

2019/01 ? spare time project

A tutorial on how to build a shiny app for the visualization of clustering results. The app helps to better identify patient data samples, e.g. during a clinical study.

Code: github

Article: Tutorial on how to build the app

Tools: R, shiny, ggplot, igraph, clustering

RTest – Pretty testing of R packages

2018/12 | Employee at Roche, Penzberg, Germany

Testing R packages can be inconvenient for people who do not know R. Matthias Pfeifer and I came up with the package RTest that enables testing R with human-readable files and generating human-readable reports.

Read more: https://towardsdatascience.com/rtest-pretty-testing-of-r-packages-50f50b135650

Tools: R, XML, HTML, CSS

Rasbperry Pi internet radio

2018/03 – spare time project

I got an old SONY radio for free. The radio was not working anymore, but the amplifier was fine. Though I thought, make it an internet radio. The radio is now controlled by the old buttons, but inside there is a raspberry pi which streams radio channels. They are visualized at an LCD screen:

Code: github

Tools: Raspberry Pi, Python, bash scripting

Analyzing genetic data for biotechnological production processes

2015/04 – 2017/08 | Employee at Insilico Biotechnology, Stuttgart, Germany

At Insilico Biotechnology I built multiple projects:

  • Network reconstruction by differential expression data with R and KNIME – Publication
  • Network reconstruction by NCA(Network component analysis) – Python, CPLEX (IBM)
  • Flux Balance Analysis with Python, cobra-toolbox and Jupyter
  • Metabolic Network reconstruction with Insilico Designer

Tools: Python (pandas, CPLEX), R (ggplot, limma), Jupyter, docker, SPARK, KNIME

Analyze last.fm with R

2017/12 – spare time project

I love listening to music. For years I have been using last.fm and analyzing the data with MATLAB. Then I thought I should try R6 instead. The outcome looks nice.

code: github

Tools: R6, RCurl, last.fm