About

Designing and coding software is what I do and what my passion is.

Software architecture and design

In a fast moving world software architectures need to be evolutionary, allowing to apply changes easier and cheaper. With that in mind I love to design systems following the domain driven design principles, to decompose functional and to utilize micro service and micro frontend approaches. Giving DevOps teams such architectures empower them to develop their sub systems autonomously, in polyglot environments and within a technological freedom to use the best tool for the job. Combined with agility it's my recipe to help developing great products and to do it more efficiently.

DevOps stack

I do professional coding in JavaScript and Java using state of the art frameworks like Node, Spring Boot, VueJS and Angular.

Deep theoretical and practical knowledge in data management and information retrieval allow me to choose the right stack for storing and processing data out of existing SQL or NoSQL databases, search and messaging engines. My favourites so far are Oracle, MySQL, Redis, MongoDb, Elasticsearch, Kafka, JMS, MQTT.

I'm a fan of operating applications using Docker, managed by Kubernetes or AWS and monitored by ELK stack and Prometheus. Immutable infrastructure and CI/CD pipelines with highly automated tests and deployments are essential for me.

Principles and further interests

My actions in finding technical solutions are characterized by a systematic analysis of the problem and requirements, the finding of possible alternatives and their assessment. I like to base decisions on figures, dates and facts.

My favorite clean code developer principles are KISS und the SOLID patterns. I love simple, elegant and performant solutions.

I'm excited about any kind of robotics, algorithms, AI and IoT.

Featured Projects

Elasticsearch plugin for linear regression

I'm the creator of a machine learning plugin for Elasticsearch providing aggregations to compute multiple linear regression on search results in real-time for predictive analytics. The implementation is open source and is based on an elaborated parallel, single-pass OLS estimation algorithm for multiple linear regression well-fitting for large-scale, distributed data sets.

Check it out

logsniffer

Inspired by tools like tail, grep and less I've created logsniffer with the vision to provide the same experience by a smart web tool when working with log files. logsniffer supports viewing all logs instantly in front of a browser, tailing logs in real-time, scrolling and searching through very large files – forward and backward.