Welcome to b-data
Data Science Services and Infrastructure
Data Science Services and Infrastructure
Take advantage of our more than 15 years of experience to make your Data Science
project a success!
Take a look at Olivier’s skills and professional experience on benz0li.b-data.io.
As advocates of OSS, we sponsor the QGIS project, the Julia language, the R Foundation and the FreeBSD Foundation.
We devote about 20% of our time to OSS and maintain
dev containers,
docker images
and
deployment templates
for Data Scientists, ML/AI Engineers, and the like 🧑💻.
We provide Data Science dev containers
for use with
VS Code (local) and
GitHub Codespaces
(web).
🎯 A unified IDE for the Data Science programming languages
R, Python,
Julia and Mojo, too.
🔥 Most images are also available in a GPU accelerated
(nvidia/cuda-based) version.
You can find all this for free on GitHub and GitLab!
Besides, see Olivier’s (benz0li’s) work on
GHC musl – Unofficial binary
distributions of GHC on Alpine Linux.
ℹ️ The docker image used to build the statically linked Linux binary releases
of Pandoc.
We are interested in your specific use case, wherefore we offer a lean and
tailor-made solution.
For a general introduction to Data Science, we recommend DataCamp.
R expertise
Databases with R (DBI,
odbc)
Data Exploration (tidyverse)
Data Visualisation (ggplot2)
Probability Distributions (actuar,
distr)
Time Series Analysis (zoo,
forecast)
Machine Learning (caret,
xgboost)
RESTful APIs (plumber, OpenCPU)
Git expertise
Reproducible Data Science with Git
Apply a successful Git branching model
Every customer project is checked-in a Git repository. Our credo: The code is
yours.
Free access to our Data Science Infrastructure with
JupyterLab and
GitLab CE included.
We help building your own Data Science software stack!
on-premise with
Debian/Ubuntu,
‘Rocky Linux’/RHEL
Virtualisation
Docker, Kubernetes
IDEs
JupyterLab + code-server
RStudio (on demand)
Check out our Jupyter demo environment at https://demo.jupyter.b-data.ch or run it locally with Docker Desktop.
docker run -it --rm -p 8888:8888 -v "${PWD}":/home/jovyan glcr.b-data.ch/jupyterlab/r/verse
Run the initial command in an empty directory so that the container populates
it.
Visit http://127.0.0.1:8888/lab?token=<token> in a browser to load
JupyterLab.
Below are the success stories of selected projects.
Stricter security policies have increasingly limited the local execution of R
and RStudio at Agroscope. As part of an architecture redesign, IT Agroscope has
decided to centralise these applications.
With RStudio Server and GitLab,
Olivier understood the complex needs of the researchers and was able to fully consider them in the integration and operational concept. Of course, the migration brought about changes, but overall this resulted in new opportunities for scientific data analysis.
– Manuel K. Schneider, Research Group Forage Production and Grassland
Years of collecting customer data can result in a valuable asset for companies.
But what to do if the data is not stored in a suitable format for future
projects, or if duplicate data affects quality?
This is precisely the question that
zkipster, a start-up in event management
software, posed to
We have very quickly gained knowledge important to our company: very specific knowledge that did not exist within zkipster, to which we had no access and for which we could not afford a full-time employee. Olivier was available at the right time, he supported us in a very professional manner and achieved great things in a short period.
– David Becker, CEO and Co-Founder
Phone number
Switzerland: +41 44 586 30 72
Business hours
Monday - Friday: 8:00 - 17:00
